Sie sind auf Seite 1von 138

NetApp Unified Storage Performance Management Using Open Interfaces

Network Appliance, Inc. March 2010

Executive Summary NetApp unified storage systems support multiprotocol data access and can be configured as SAN, IPSAN and NAS devices at the same time. NetApp storage systems support different type of storage objects like aggregates, volumes, LUNs, qtrees etc. NetApp provides open interfaces like Data ONTAP APIs, SNMP, SMI-S agent for monitoring and managing various components of the NetApp storage system. This document provides the details of how to use NetApp open interfaces for unified storage performance management and how to simplify performance management of NetApp storage systems when multiple protocols are supported and multiple objects are being managed

Table of Contents
1 INTRODUCTION............................................................................................................................... 3 1.1 1.2 1.3 2 BACKGROUND .............................................................................................................................. 3 UNIFIED PERFORMANCE MANAGEMENT ...................................................................................... 3 PURPOSE AND SCOPE.................................................................................................................... 3

NETAPP OPEN INTERFACES........................................................................................................ 5 2.1 2.2 2.3 Data ONTAP APIs ...................................................................................................................... 5 SNMP .......................................................................................................................................... 5 DATA ONTAP SMI-S AGENT ...................................................................................................... 6

NETAPP PERFORMANCE MANAGEMENT CONCEPTS......................................................... 7 3.1 3.2 3.3 BASICS OF PERFORMANCE ANALYSIS .......................................................................................... 7 PERFORMANCE MONITORING INFRASTRUCTURE .......................................................................... 7 PERFORMANCE COUNTERS DESIGN .............................................................................................. 9

PERFORMANCE MEASUREMENT USING NETAPP OPEN INTERFACES ....................... 10 4.1 4.2 4.3 4.4 USING AGGREGATE COUNTERS .................................................................................................. 10 USING LUN COUNTERS .............................................................................................................. 11 USING NFSV3/CIFS COUNTERS................................................................................................. 12 USING PROCESSOR/SYSTEM COUNTERS ..................................................................................... 13

APPENDIX 1 - PERFORMANCE COUNTERS IN DATA ONTAP 7.3..................................... 16 5.1 5.2 5.3 5.4 COUNTER TYPES ........................................................................................................................ 16 COUNTER PRIORITY LEVELS ...................................................................................................... 17 LIST OF OBJECTS SUPPORTED ..................................................................................................... 18 LIST OF COUNTERS SUPPORTED IN VARIOUS OBJECTS ................................................................ 20

Performance Management Design Guide

2 of 138

3/7/2008

1 Introduction
1.1 Background

Performance measurement is important in understanding the usage patterns of storage which can help in catching the performance problems before they impact the users and applications. Usually IT infrastructures often grow rapidly beyond the initial sizing estimates. To cater to these growths as administrators add workload demand from new users and applications, the storage system latency increases linearly but at certain threshold latency starts to increase exponentially and results in what is known as the hockey stick effect. By monitoring performance and analyzing trends, administrators are able to anticipate the latency curve before it goes up and prevent it from impacting users. Performance Measurement is useful in the following areas: 1. Capacity planning and sizing 2. Tuning Storage, Platform and Application configurations to achieve the best performance 3. Monitoring health of the system through trending 4. Monitoring thresholds and alerts to avoid the bottlenecks 5. Load balancing for optimal usage of the resources 1.2 Unified Performance Management

Unified Storage capabilities of NetApp storage systems require a different approach in managing the performance of the storage system. The same storage object can be accessed through different protocols. Also there are many different storage objects, both physical and logical, which form the storage hierarchy of the system and thus complicate the performance calculation for the system. Thus a unified view of performance of the system is required to manage the performance of the NetApp storage systems. 1.3 Purpose and Scope

NetApp provides following Open Interfaces to remotely manage the NetApp devices: NetApp Manageability SDK SNMP Data ONTAP SMI-S Agent This document provides details of different performance management scenarios in the NetApp storage systems and how to use NetApp open interfaces to manage these Performance Management Design Guide 3 of 138 3/7/2008

scenarios. This document covers only the use of NetApp Manageability SDK interfaces for calculating performance details because among the open interfaces, currently only Data ONTAP APIs provide the interface for reading the performance related data.

Performance Management Design Guide

4 of 138

3/7/2008

2 NetApp Open Interfaces


2.1 Data ONTAP APIs

The ONTAPI interface is a set of foundational APIs for managing NetApp Storage Systems. This interface is developed by NetApp for advance management of NetApp Storage Systems. ONTAPI interfaces use XML as a format and may be configured to use either HTTP, HTTPS or DCE/RPC as the transport mechanism. HTTP is important when managing devices which may be outside of the corporate firewall, and locked down so that only port 80 is available, and is the default transport configuration. HTTPS can be used for secure communication through ONTAPI. NetApp Manageability SDK also provides a small set of core interfaces that marshal and un-marshal ONTAP API arguments using XML as the description language. At present, core interfaces are provided in C/C++, Perl and Java.

2.2

SNMP

SNMP (Simple Network Management Protocol) is a well-known standard for network management. NetApp storage systems support the SNMP version 1 compatible agent. This agent supports both MIB-II and the NetApp custom MIB. For reasons of security, NetApp supports only monitoring using SNMP which means that SNMP SET operations are not permitted. If SNMP is enabled in Data ONTAP, SNMP managers can query your storage system's SNMP agent for information (specified in your storage system's MIBs or the MIB-II specification). In response, the SNMP agent gathers information and forwards it to the SNMP managers using the SNMP protocol. The SNMP agent also generates trap notifications whenever specific events occur and sends these traps to the SNMP managers. The SNMP managers can then carry out actions based on information received in the trap notifications. The latest versions of the Data ONTAP MIB files are available online on the NetApp on the Web (NOW) site.

Performance Management Design Guide

5 of 138

3/7/2008

2.3

Data ONTAP SMI-S Agent

Data ONTAP SMI-S agent provides standards based storage management interface to discover, monitor, and manage NetApp storage systems. The specifications for SMI-S are developed by SNIA (Storage Networking Industry Association) and DMTF (Distributed Management Task Force) standards organizations. The SMI-S agent is implemented as a proxy-based management solution. The agent needs to be installed on an external server (i.e. it is not implemented within ONTAP). Currently the supported platforms for SMI-S agent installation are Windows and Linux (Red Hat and SUSE) based hosts.

Performance Management Design Guide

6 of 138

3/7/2008

3 NetApp Performance Management Concepts


3.1 Basics of Performance Analysis
One of the fundamental theories used for Performance Analysis is Littles Law. Following is the mathematical representation of Littles Law: Queue-Length = Throughput x Latency Throughput - The rate at which any request coming into the system is served Latency The time taken to service a request coming into the system Queue-Length Average number of pending requests in the system at any given time Littles Law can be used to analyze the performance of any system that can be modeled as queues. For example, to analyze how efficiently a Bank is servicing its customers, we need to calculate at what rate the customers are entering the bank for a service (Throughput measurement) and how much time the teller at the counter takes to service each customer (Latency measurement). Once we get Throughput and Latency data, we can calculate the average length of the customer queue which is a direct measure of how efficiently the bank is servicing its customers and this data can be used to take a decision whether the bank needs to open one more counter or the employee at the counter needs to be made more efficient. Littles Law can be used to understand the performance of any storage component in the customer environment. For each component in the storage system, the queue-length data can be calculated using the I/O rate to the component (gives throughput measurement) and the I/O latency of the component. For each environment, the queue-length numbers need to be monitored and the average queue-length should be bench-marked. The performance hot-spots can be identified in the specific storage environment by tracking the consistent periodic peaks in the queue-length numbers observed over a period of time. 3.2 Performance Monitoring Infrastructure

The following figure shows the overall infrastructure available for applications to monitor the performance statistics of NetApp storage systems.

Performance Management Design Guide

7 of 138

3/7/2008

3rd Party Management Tools

Autosuppo rt

DFM Windows Perfmon Clients CLI

Filer Data Archives

Data ONTAP APIs SNMP

SMB calls Windows Perfmon Support

Existing Performanc e Commands

Ontap Counter Manager (CM)

Storage System Performance counters

At the base of the performance infrastructure there are a set of performance counters accessible through existing performance commands. Above this set of performance counters, is a thin layer of abstraction called Counter Manager (CM). Counter Manager provides an object abstraction to higher layers a single view of ONTAP performance counters a standard performance API set for all the clients of CM The following are the different channels through which performance monitoring applications can access the API set of the CM: CLI commands o Stats o Sysstat o statit Windows Perfmon clients using SMB calls to Windows Perfmon support Data ONTAP APIs o Used for building NetApp management tools like DFM-Performance Analyzer, PerfStat, PerfViewer o Also used by third parties to integrate their performance management tools for NetApp storage systems. SNMP In future we will be adding something called Filer Data Archives. Today, when you read performance counters you get a real-time view of these counters. You can save them but

Performance Management Design Guide

8 of 138

3/7/2008

that responsibility is up to the host or the administrator. With Filer Data Archives, the storage system keeps historical versions of performance counters so that it's easy to do trending and analysis without the host having to save all that information. 3.3 Performance Counters Design

Performance counters are grouped in an object-instance-counter hierarchy. The object will have a set of well defined counters. Each object could have many instances of it but each of these instances will have its own corresponding values for this well defined counter set. For example, Volume is an Object type. In a particular storage system, there could be many volumes, for instance vol0 and vol1. Each of these volumes is an instance of the Volume object. The Volume object has a well defined counter set that includes counters like read_ops, read_latency etc. Each instance of the Volume object in this storage system viz. vol0 and vol1 have its own values for the counters read_ops, read_latency etc. The example is illustrated below:

vol0

read_ops reads_latency

Volume
vol1
read_ops reads_latency

Each counter has a type, unit and priority associated with it. See Appendix 1 Performance counters in DataONTAP 7.1 for the description of the various performance objects and the associated counters supported in DataONTAP 7.1, meaning of the different types and priorities of the counters.

Performance Management Design Guide

9 of 138

3/7/2008

4 Performance Measurement using NetApp Open Interfaces


The following sections describe how the performance counters associated with some of the performance objects can be used in various scenarios that would need performance measurements.

4.1

Using Aggregate counters

Consider the scenario wherein the application needs to know if more disks need to be added to the aggregate to improve the application performance. For this the application needs to do the following: Get the number of disks in the aggregate Get the total transfer rate of IOPs to the aggregate Get the rate of average block/byte transfer rate to the aggregate. The following table provides details on API calls to get the above data: Operation Get the number of disks in the aggregate Get the Total IOP rate and average block transfer rate Command Details (Data ONTAP API) aggr-list-info Comments API returns aggr_info[] array. The diskcount attribute in the array provides the number of disks in the aggregate. Make two calls to read the data at two points in time: t1 and t2, and calculate the throughput during the interval

perf-object-get-instances The input to this command is to be set as below: Object Name : aggregate Instance Name : aggregate name of interest Counter Info : Fill three elements of type counter_infowith the counter names as total_transfers, user_read_blocks and user_write_blocks respectively

Assuming that the data in the aggregate are striped across all the disks equally and that the benchmark numbers for the FC disks are: Throughput: 30 50 MB/sec, and IOPS: 200 400

Also assume that the application is using 64K size blocks for read/write operations.

Performance Management Design Guide

10 of 138

3/7/2008

Total_transfers rate = (total_transfers at t2 - total_transfers at t1)/ (t2 t1) Total MB/sec of read/write data = ((user_read_blocks + user_write_blocks at t2) - (user_read_blocks + user_write_blocks at t1))*64/ ((t2 t1)*1024)

IOPS per disk = (total_transfers rate)/(disk_count) Throughput per disk = (Total MB/sec of read/write data) / (disk_count) If the IOPS or the throughput per disk is not in the range mentioned above then additional disks should be added to the aggregate to improve the performance of application that is doing IOs to that particular aggregate. 4.2 Using LUN counters Consider the scenario where an application is sending I/O packets to a LUN and the response time of the I/O is poor. This could be because of one or more of the following reasons: Poor concurrency design of the application measured by the number of threads of execution in the application Slower LUN measured by the average latency of the LUN Busy LUN measured by the number of queue-full messages per second If the application concurrency is taken care with good multi-threaded design then the bottle-neck is most probably at the LUN. To diagnose the LUN bottleneck, do the following: Check the average latency of the LUN. Assuming that the application is running in Oracle/Exchange environment , the threshold latency is around 20 millisecond Check queue full message rate of the LUN. An unusually high rate of queue full messages indicates the LUN is not able to handle the I/O rates of the application. The following table provides details on API calls to get the above data: Operation Command Details (Data ONTAP API) perf-object-get-instances Get the LUN counters avg_latency and queue_full The input to this command is to be set as below: Object Name : lun Instance Name : LUN name of interest Counter Info : Fill three elements of type counter_info with the counter names Comments Get the data at two intervals of time t1 and t2

Performance Management Design Guide

11 of 138

3/7/2008

avg_latency, total_ops and queue_full Queuefull/sec = (queuefull rate at time t2 - queuefull rate at t1) / (t2 t1) avg_latency = (avg_latency at time t2 - avg_latency at time t1) / (total_ops at time t2 total_ops at time t1) From the above calculations, the application can determine if the queuefull rate and average latency are in the acceptable range. If the queue full rate is not in the acceptable range then the LUN is really busy, add more LUNs and distribute the I/O to different LUNs. If the average latency is not in the acceptable range, then the LUN is slow. Move the LUN to the volume which is based on a faster disk. If the latency at the server is much more compared to the average latency of the LUN then increase the queue-depth settings of the SCSI/FC driver on the host. 4.3 Using NFSV3/CIFS counters Consider the scenario, where an application reading data from NFS/CIFS share is facing timeouts or disconnects. The problem could be due to network latency or storage system latency. Following are the steps to debug this issue: Look on storage system for high latencies. The NFS and CIFS read latency can be calculated as shown below: Operation Command Details (Data ONTAP API) Comments Get the data at two intervals of time t1 and t2

perf-object-get-instances Get the nfsv3 counters nfsv3_read_latency and The input to this command is to be set as nfsv3_avg_read_latency_base below: Object Name : nfsv3 Instance Name : NFS share name of interest Counter Info : Fill two elements of type counter_info with the counter names as nfsv3_read_latency and nfsv3_avg_read_latency_base respectively perf-object-get-instances Get the cifs counters cifs_latency and The input to this command is to be set as cifs_latency_base below: Object Name : cifs Instance Name : CIFS share name of interest Counter Info : Fill two elements of type

Get the data at two intervals of time t1 and t2

Performance Management Design Guide

12 of 138

3/7/2008

counter_info with the counter names as cifs_latency and cifs_latency_base respectively Calculation of the Latencies for NFS and CIFS shares: NFS read latency = (nfsv3_read_latency at time t2 - nfsv3_read_latency at time t1) / (nfsv3_avg_read_latency_base at time t2 - nfsv3_avg_read_latency_base at time t1) CIFS latency = (cifs_latency at time t2 - cifs_latency at time t1) / (cifs_latency_base at time t2 - cifs_latency_base at time t1) The maximum acceptable latencies are: 60 msecs for NFS 30 msecs for CIFS 120 msecs for FCP/iSCSI If the latencies as calculated above are higher than the maximum acceptable latencies, then the problem lies with the storage system If storage system doesnt show high latencies, this is probably a network issue. Compare host view of latencies versus storage system view of latencies. If different, this is probably a network or host I/O driver issue.

4.4

Using Processor/System counters

One of the unique features of NetApp storage systems is, unlike other vendor storage systems, NetApp storage systems have an OS and File System. So CPU of the storage box could also be one of the potential candidates for bottlenecks in the I/O path. Consider the scenario where there is high latency and application sluggishness. Assume that there are no disk bottlenecks or application concurrency problems. Then the problem could be with the utilization of the CPUs on the storage system. Following are the steps to diagnose the CPU bottleneck problems: Look for CPU utilizations of more than > 95% on 1P, >180% on 2P, >350% on 4P If the CPU utilization is not high enough (i.e., the utilization is less than the above mentioned ranges), then look for the total domain utilizations of more than 90%. Improper management of storage system could result in all work that needs to be done is in one domain The CPU and domain utilizations can be calculated as described below. The following table shows how to use APIs to get the required counters to calculate these values:

Performance Management Design Guide

13 of 138

3/7/2008

Operation Get the System object counters total_processor_busy, cpu_elapsed_time

Command Details (Data ONTAP API) perf-object-get-instances The input to this command is to be set as below: Object Name : system Instance Name : system Counter Info : Fill two elements of type counter_info with the counter names as total_processor_busy and cpu_elapsed_time respectively perf-object-get-instances The input to this command is to be set as below: Object Name : processor Instance Name : (keeping this empty will result in getting the counter values for all of the processors in the system in one shot) Counter Info : Fill two elements of type counter_info with the counter names as processor_elapsed_time and domain respectively

Comments Get the data at two intervals of time t1 and t2

Get the Processor object counters domain and processor_elapsed_time for each of the processor instance

Get the data at two intervals of time t1 and t2 The domain counter will result in multiple arrays of processor time spent in various domains. The number of arrays is equal to the number CPUs in the storage system

% CPU utilization = (total_processor_busy at time t2 - total_processor_busy at time t1) / (cpu_elapsed_time at time t2 - cpu_elapsed_time at time t1) Domain utilization: For each processor instance, the domain utilization is got as below: %domain utilization = (domain[cpu-number][domain element] at time t2) (domain[cpunumber][domain element] at time t1) / (processor_elapsed_time at time t2 processor_elapsed_time at time t1) The values may look as below: Idle Kah Net Stor Exem Raid Targ Netc Netc2

Performance Management Design Guide

14 of 138

3/7/2008

Processor 0: 20% 34% 20% 10% 6% Processor 1: 23% 60% 10% 5% 1% Total

10% 0% 0% 0% 1% 0% 0% 0%

43% 94% 30% 15% 7% 11% 0% 0% 0%

The total domain utilization is got by adding the corresponding elements in the above arrays. Once the % CPU utilization and % domain utilization is calculated as above, the solution for the problem could be decided as below: 1. If the CPU utilization is beyond the threshold values as mentioned above, then the CPU bottleneck could be solved by moving to a storage controller with more processors 2. If the CPU utilization is not high enough and one of the domain elements utilization is more than the threshold values as mentioned above, then the bottleneck problem is solved by proper prioritization of the workloads using flexshare.

Performance Management Design Guide

15 of 138

3/7/2008

5 Appendix 1 - Performance counters in Data ONTAP 7.3


The following list may not be exhaustive or the most recent. The actual list of Performance counters that is applicable to your storage system can be found by using the ONTAP APIs perf-object-list-info and perf-object-counter-list-info on your storage system.

5.1 Counter Types


Type RAW RATE Meaning Means that the counter being exported has no math associated with it. Means that two samples of this counter (say C1 and C2) have to be taken at two different points in time (say t1 and t2) and the following math needs to be worked out: (C2 - C1) / (t2 t1) Means that two samples of this counter (say C1 and C2) have to be taken at two different points in time (say t1 and t2) and the following math needs to be worked out: (C2 - C1) Means that two samples of this counter (say C1 and C2) and two samples of the base counter (say b1 and b2) have to be taken at two different points in time (say t1 and t2) and the following math needs to be worked out: (C2 - C1)/(b2-b1) Means that two samples of this counter (say C1 and C2) and two samples of the base counter (say b1 and b2) have to be taken at two different points in time (say t1 and t2) and the following math needs to be worked out: 100 * (C2 - C1)/(b2-b1) Means that this counter is a string Typically a base counter that is needed for math associated with other counters. These counters are never exported or displayed

DELTA

AVERAGE

PERCENT TEXT NODISP

Performance Management Design Guide

16 of 138

3/7/2008

5.2 Counter Priority Levels

Level BASIC ADMIN ADVANCED DIAG

Meaning

Essential counters for problem diagnosis Counters for administrative purposes Advanced counters Need not be customer visible Counters that used for diagnosis purposes. Some of these counters can report incorrect data too! PERFMON_* Counters that are visible through perfmon. Counters that are visible through the CLI with 'stats'. STATS_* Counters that are visible through ZAPI. ZAPI_* Counters that are visible through autosupport. ASUP_* Objects that are automatically sampled. SAMPLED

Note: The legacy counter priority levels (BASIC-DIAG) are by default visible to all consumers. Specific consumers can be included by logically ORing their flag, or excluded by negating their flag. (CM_BASIC | CM_SAMPLED) indicatse an object includes the 'sampled per second bit', while (CM_BASIC | ~CM_PERFMON_BASIC) indicates a counter is not visible for Perfmon. The priority level attached with a counter is just a suggestion to various clients of CM regarding the importance/correctness of a counter. It is up to the clients to do any filtering based on these priorities.

Performance Management Design Guide

17 of 138

3/7/2008

5.3 List of Objects supported


Object Name aggregate audit_ng bufcache cacheeject cachequeue
CIFS

Contains CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_PRIV_DIAG_ZAPI | CM_CP_DUMP CM_PRIV_DIAG_ZAPI | CM_CP_DUMP CM_PRIV_DIAG_ZAPI | CM_CP_DUMP CM_PRIV_DIAG_ZAPI | CM_CP_DUMP CM_PRIV_DIAG_ZAPI | CM_CP_DUMP CM_BASIC CM_BASIC|CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_BASIC CM_DIAG CM_DIAG CM_ADVANCED CM_BASIC CM_BASIC CM_BASIC 18 of 138 3/7/2008

cifs_ops cifs_session_signing cifs_stats cifs_watch cifsdomain cpdump cpdump_disk cpdump_fvol cpdump_rg cpdump_rvol disk
dump

ext_cache ext_cache_obj FCP flexcache fpolicy_global fpolicy_stats_policy fpolicy_stats_server hostadapter ifnet incpr_global incpr_sess iomem
iSCSI LDAP

lmem lmgr_ng logical_replication_destination logical_replication_source LUN Performance Management Design Guide

NDMP nfsv3 nfsv4 nrv NVRAM partial_file_restore perf priorityqueue prisched processor qtree quota raid raid_stripe readahead replication rsm_src_global rsm_dst_global rsm_src_relation rsm_dst_relation SIS Sparse Spinhi System tape test vfiler volume volume_snapmirror_destination volume_snapmirror_source vscan_server_stat vscan_stats wafl cifs_watch zodiac

CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_ADVANCED CM_ADVANCED CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC CM_BASIC CM_DIAG CM_TEST CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

19 of 138

3/7/2008

5.4 List of Counters supported in various Objects


aggregate object Counter Name parent_host total_transfers user_reads user_writes cp_reads user_read_blocks user_write_blocks cp_read_blocks wv_fsid wv_vol_type wv_fsinfo_fs_versio n wv_volinfo_fs_optio ns wv_volinfo_fs_flags wv_fsinfo_blks_rese rve wv_fsinfo_blks_use d wv_fsinfo_blks_use d_by_plane0 wv_fsinfo_blks_blks _rsrv_holes_cifs wv_fsinfo_blks_blks Size 64B 8B 8B 8B 8B 8B 8B 8B 16B 16B 4B 16B 16B Type STRING RATE RATE RATE RATE RATE RATE RATE STRING STRING RAW STRING STRING RAW RAW RAW RAW RAW RAW Display Unit NONE PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE Description Name of parent host Total number of transfers per second serviced by the aggregate Number of user reads per second to the aggregate Number of user writes per second to the aggregate Number of reads per second done during a CP to the aggregate Number of blocks read per second on the aggregate Number of blocks written per second to the aggregate Number of blocks read per second during a CP on the aggregate File system ID for the aggregate Volume type for the aggregate File system version for the aggregate File system options for the aggregate File system flags for the aggregate Total blocks in the aggregate Reserved blocks in the aggregate Used blocks (all planes) in the aggregate Used blocks (plane 0) in the aggregate Reserved blocks for CIFS holes in the aggregate Reserved blocks for all holes in Priority CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depen ds on

wv_fsinfo_blks_total 8B 8B 8B 8B 8B 8B

Performance Management Design Guide

20 of 138

3/7/2008

_rsrv_holes wv_fsinfo_blks_blks _rsrv_overwrite wv_fsinfo_blks_rsrv _absents wv_fsinfo_blks_sna p_reserve_pct wv_fsinfo_blks_res_ state 8B 8B 4B 4B RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE NONE PERCT NONE PERCT NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

the aggregate Overwrite reserved blocks in the aggregate Absent sparse volume reserved blocks in the aggregate Snap reserved percentage for the aggregate Space reservation state for the aggregate Overwrite slider percentage for the aggregate Total inodes in the aggregate Reserved inodes in the aggregate Used inodes in the aggregate CP when block reallocation first performed Blocks past EOF in the aggregate Snap reserved blocks in the aggregate Blocks used in zombie files in the aggregate Total delalloc blocks from flexible volumes in the aggregate Total reserved delalloc blocks from flexible volumes in the aggregate Total blocks for holes from flexible volumes in the aggregate Total blocks for overwrites from flexible volumes in the aggregate Total blocks for always overwrites from flexible volumes in the aggregate Total indirect blocks of all container files Active file system freed blocks during CP in the aggregate CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

wv_fsinfo_blks_over 4B write_slider_pct wv_fsinfo_inos_total 4B wv_fsinfo_inos_rese rve wv_fsinfo_inos_use d wv_fsinfo_blkr_cp wvblk_past_eof wvblk_snap_reserve wvblk_zombie_blks wvblk_child_delallo c 4B 4B 4B 8B 8B 8B 8B

wvblk_child_rsrv_de 8B lalloc wvblk_rsrv_child_h oles wvblk_rsrv_child_o verwrite wvblk_rsrv_child_o verwrite_always wvblk_child_indirec t_blk_cnt wvbd_active_frees 8B 8B 8B 4B 8B

Performance Management Design Guide

21 of 138

3/7/2008

wvbd_whole_frees wvbd_active_frees_ y wvbd_owner_chang ed_y wvbd_whole_frees_ o wvblk_saved_fsinfo _inos_total wvblk_saved_fsinfo _inos_reserve wvblk_saved_fsinfo _inos_used wvblk_delalloc wvblk_rsrv_delalloc wvblk_rsrv_holes_ci fs wvblk_rsrv_holes wvblk_rsrv_overwrit e disk_type wvdf_num_frees wvdf_max_frees wvblk_child_to_be_ reclaimed blkr_async_offline

8B 8B

RAW RAW

NONE NONE

Active file system whole freed blocks during CP in the aggregate Active file system freed blocks (youngest) during CP in the aggregate Changed ownership blocks (youngest) during CP in the aggregate

CM_DIAG CM_DIAG

8B 8B 4B 4B 4B 4B 4B 4B 4B 4B 4B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW DELTA

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG

Whole freed blocks (oldest) during CM_DIAG CP in the aggregate Total inodes during CP in the aggregate Reserved inodes during CP in the aggregate Used inodes during CP in the aggregate Delalloc blocks during CP in the aggregate Reserved delalloc blocks during CP in the aggregate Blocks reserved for CIFS holes during CP in the aggregate Blocks reserved for all holes during CP in the aggregate Blocks reserved for overwrite during CP in the aggregate Primary type of disk used for the aggregate Total number of delayed pvbn frees in the aggregate Maximum number of delayed pvbn frees in the aggregate Total Outstanding delayed pvbn frees in the aggregate Number of times an async block reallocation call failed due to aggr offline Number of times an async block reallocation call failed due to no message 22 of 138 CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

blkr_async_no_msg

8B

DELTA

NONE

CM_DIAG

Performance Management Design Guide

3/7/2008

blkr_async_no_mem

8B

DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

Number of times an async block reallocation call failed due to no memory Number async block reallocation requests launched Number async block reallocation requests completed Total number of blocks scanned for block reallocation Number of free blocks scanned for block reallocation Number of super blocks scanned for block reallocation Number of blocks scanned in current Allocation Area Number of aggregate snapshot blocks scanned for block reallocation Total number of segments scanned Number of segments skipped because segment full Number of segments skipped because segment empty Number of segments skipped because of policy rejection

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

blkr_async_launched 8B blkr_async_complet ed 8B

blkr_blocks_scanned 8B blkr_free_blocks_sc anned blkr_super_blocks_s canned blkr_aa_blocks_scan ned 8B 8B 8B

blkr_aggrsnap_block 8B s_scanned blkr_segments_scan ned blkr_full_segments_ scanned blkr_empty_segment s_scanned blkr_rejected_segme nts_scanned blkr_rejected_blocks _scanned blkr_reads_launched blkr_blocks_read blkr_blocks_dummy _read 8B 8B 8B 8B 8B 8B 8B 8B

Number of blocks skipped because CM_DIAG of policy rejection Number of read I/Os launched for block reallocation Number of blocks read for block reallocation Number of dummy blocks read for block reallocation CM_DIAG CM_DIAG CM_DIAG

blkr_blocks_postfilte 8B red blkr_blocks_overwri tten 8B

Number of read blocks postfiltered CM_DIAG during block reallocation Number of blocks skipped because of overwrite during block CM_DIAG reallocation

Performance Management Design Guide

23 of 138

3/7/2008

blkr_blocks_realloca ted blkr_blocks_redirect ed blkr_blocks_redirect ed_maybe blkr_blocks_redirect ed_noread blkr_blocks_redirect ed_nol1 blkr_blocks_redirect ed_noio blkr_blocks_redirect ed_reread blkr_blocks_redirect ed_noverify blkr_redirect_susps blkr_redirect_ra_ma p blkr_redirect_ra_l0 blkr_redirect_kireeti s_scanned blkr_redirect_ra_l1 blkr_redirect_deman d_req blkr_redirect_deman d_rereq blkr_redirect_deman d_drop

8B 8B 8B 8B

DELTA DELTA DELTA DELTA

NONE NONE NONE NONE

Number of blocks reallocated during block reallocation Number of blocks requiring redirection Number of blocks with possible pvbn matches (may avoid read) Number of blocks requiring redirection not read because pvbns matched Number of blocks requiring redirection because container L1 not available Number of blocks requiring redirection because RAIDio not available Number of blocks reread because of no L1 or no RAIDio Number of blocks read without checksum verification Number of suspensions for redirect scan worker Number of map blocks read ahead during redirect scan Number of container L0s read ahead during redirect scan Number of kireeti blocks scanned during redirect scan Number of container L1s read ahead during redirect scan Number of on demand redirect update requests Number of on demand redirect update requests for blocks already enqueued Number of on demand redirect update requests dropped Number of indirect blocks inspected Number of indirect blocks not needing updated 24 of 138

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

8B

DELTA

NONE

CM_DIAG

8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B

DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

blkr_redirect_indirec 8B ts_inspected blkr_redirect_indirec 8B ts_ok

Performance Management Design Guide

3/7/2008

blkr_redirect_indirec 8B ts_updated blkr_redirect_blocks _updated blkr_redirect_blocks _invalid blkr_redirect_blocks _ok blkr_policy1_reject_ reasons wvzmb_num_zmsgs _inuse wvblk_space_tax 8B 8B 8B 8B 4B 8B

DELTA DELTA DELTA DELTA DELTA RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

Number of indirect blocks not needing updated Number of blocks updated Number of blocks invalid (not updated) Number of blocks ok Number of segments rejected per reason by policy1 Number of zombie/sfsr messages working Total space tax for all volumes in the aggregate Total count of lev0 container blocks over nominal for all volumes Count of indirect blocks that are delalloc'd Count of blocks in the delete log Count of indirect blocks in the delete log Count of blocks in the purge file Count of indirect blocks in the purge file Containment version for SnapLock Compliance volume Containment version for SnapLock Enterprise volume Containment version reserved for SnapLock Containment version reserved for SnapLock Highest containment version detected for SnapLock Compliance volume Highest containment version detected for SnapLock Enterprise volume Highest containment version

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

wvblk_lev0_over_no 8B minal wvblk_ind_delalloc dlog_blks_total dlog_blks_indirect purge_blks_total purge_blks_indirect wv_fsinfo_containm ent_version_slc wv_fsinfo_containm ent_version_sle wv_fsinfo_containm ent_version_spare1 wv_fsinfo_containm ent_version_spare2 8B 8B 8B 8B 8B 4B 4B 4B 4B

wv_fsinfo_containm ent_version_highest_s 4B lc wv_fsinfo_containm ent_version_highest_s 4B le wv_fsinfo_containm 4B

RAW RAW

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

25 of 138

3/7/2008

ent_version_highest_s pare1 wv_fsinfo_containm ent_version_highest_s 4B pare2 RAW NONE

reserved for SnapLock Highest containment version reserved for SnapLock

CM_DIAG

audit_ng object Counter Name event_throughput Size 8B Type RATE Display Unit Description Priority CM_DIAG Depends on Amount of audit KB_PER_SEC information recorded to disk per second MSECS Average latency for recording an event to disk Number of events recorded to disk per second Average latency for determining whether or not to audit Number of SpinNP operations monitored Number of SpinNP operations recorded to disk Amount of time since the consolidator last polled (vserver only) Amount of active staged data waiting to be consolidated (vserver only)

event_latency

8B

AVERAGE

CM_DIAG

events

events

8B

RATE

PER_SEC

CM_DIAG monitored_sp innp_ops

spinnp_determination 8B _latency monitored_spinnp_op s recorded_spinnp_ops 8B 8B

AVERAGE RAW RAW

MSECS NONE NONE

CM_DIAG CM_DIAG CM_DIAG

time_since_last_poll

4B

RAW

SECS

CM_DIAG

staged_data_size

8B

RAW

NONE

CM_DIAG

consolidation_throug hput max_sacl_size

8B

RATE

Amount of active staged data KB_PER_SEC consolidated per second (vserver only) NONE

CM_DIAG

4B

RAW

Maximum SACL size CM_DIAG used during SpinNP

Performance Management Design Guide

26 of 138

3/7/2008

determination (vserver only) max_event_size 4B RAW NONE Maximum event size recorded to disk (vserver only) CM_DIAG

bufcache object Counter Name buf_count buf_empty buf_valid buf_dirty buf_read buf_write buf_cpio buf_recycle buf_norecycle buf_stolen buf_ec_inflight mem_per_buf Size 4B 4B 4B 4B 4B 4B 4B 4B 4B 4B 4B 4B Type RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW Display Unit NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE Description Buffer cache size Empty (unused) buffers Valid buffers Buffers with modified data Buffers being read Buffers being written Buffers associated with CP IO Buffers in a recycle queue Buffers loaned to other subsystems Buffers in flight to extended cache Memory used per buffer Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

Buffers that must not be recycled CM_DIAG CM_DIAG CM_DIAG CM_DIAG

cacheeject object Counter Name level0 level1 level2 level3 level0_rate level1_rate level2_rate Size 8B 8B 8B 8B 8B 8B 8B Type RAW RAW RAW RAW RATE RATE RATE Display Unit NONE NONE NONE NONE PER_SEC PER_SEC PER_SEC Description Level 0 blocks Level 1 blocks Level 2 blocks Level 3 blocks Level 0 blocks eject rate Level 1 blocks eject rate Level 2 blocks eject rate Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

Performance Management Design Guide

27 of 138

3/7/2008

level3_rate

8B

RATE PER_SEC

Level 3 blocks eject rate

CM_DIAG

cachequeue object Counter Name Size priority buf_count count_busy gone scavenge vscavenge eject_times age_out upgrade touched added removed scavenge_rate vscavenge_rate 4B 4B 4B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RATE RATE Display Unit NONE NONE NONE NONE NONE NONE MSECS NONE NONE NONE NONE NONE PER_SEC PER_SEC Description queue priority Buffers on queue Buffers on queue that are busy Queue was already marked empty in search Number of buffers scavenged Number of buffers vscavenged Total eject times when scavenged Number of buffers aged out Number of buffers upgraded Number of buffers touched Number of buffers added to queue Number of buffers removed from queue Number of blocks scavenge from this queue per second Number of blocks vscavenged from this queue per second Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

cifs object Counter Name cifs_ops cifs_op_count cifs_op_pct Size 8B 8B 8B Type RATE DELTA PERCT Display Unit PER_SEC NONE PERCT Description Total number of CIFS operations per second Array of select CIFS operation counts Array of select CIFS operation counts as a percentage of total CIFS operations Priority CM_BASIC CM_DIAG CM_DIAG cifs_ops Depends on

Performance Management Design Guide

28 of 138

3/7/2008

cifs_latency

8B

AVERA GE DELTA NODIS P DELTA RATE RATE AVERA GE AVERA GE DELTA DELTA DELTA DELTA

MSECS DELTA NODISP NONE PER_SEC PER_SEC MSECS MSECS NONE NONE NONE NONE

Average latency for CIFS operations in milliseconds Total observed CIFS operations to be used as a base counter for CIFS average latency calculation Histogram of latency for CIFS operations in milliseconds Total number of CIFS read operations per second Total number of CIFS write operations per second Average latency for CIFS read operations in milliseconds Average latency for CIFS write operations in milliseconds Histogram of latency for CIFS read operations in milliseconds Histogram of latency for CIFS write operations in milliseconds Histogram of cifs read sizes Histogram of CIFS write sizes

CM_BASIC

cifs_latenc y_base

cifs_latency_base

8B

CM_BASIC

cifs_latency_hist cifs_read_ops cifs_write_ops cifs_read_latency cifs_write_latenc y cifs_read_latency _hist cifs_write_latenc y_hist cifs_read_size_hi sto cifs_write_size_h isto

8B 8B 8B 8B 8B 8B 8B 4B 4B

CM_DIAG CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

cifsdomain object Counter Name netlogon_latency Size 8B Type Display Unit Description Priority CM_DIAG Depends on netlogon_ latency_base

Average latency for AVERAGE MSECS netlogon operations in milliseconds Total time spent waiting for netlogon requests to be used as a base counter for netlogon average latency calculation Average latency for

netlogon_latency_base 8B

DELTA

NONE

CM_DIAG

lsa_latency

8B

AVERAGE MSECS

CM_DIAG lsa_latency_base

Performance Management Design Guide

29 of 138

3/7/2008

lsa (local security authority) operations in milliseconds Total time spent waiting for lsa requests to be used as a base counter for lsa average latency calculation

lsa_latency_base

8B

DELTA

NONE

CM_DIAG

samr_latency

8B

Average latency for samr (security account AVERAGE MSECS manager RPC service) operations in milliseconds Total time spent waiting for samr requests to be used as a base counter for samr average latency calculation

CM_DIAG samr_latency_base

samr_latency_base

8B

DELTA

NONE

CM_DIAG

cifs_ops object Counter Name get_attr_ops set_attr_ops get_attr_ext_ops Size 8B 8B 8B Type RAW RAW RAW Display Unit NONE NONE NONE Description Query Information operations (SMB Code = 0x08) Set Information operations (SMB Code = 0x09) Query Information2 operations (SMB Code = 0x23) Set Information2 operations (SMB Code = 0x22) Query FS Information operations (SMB Code = 0x32SubCode = 0x03) Query Path Information operations (SMB Code = 0x32SubCode = 0x05) Priority CM_DIAG CM_DIAG CM_DIAG Depends on

set_attr_ext_ops

8B

RAW

NONE

CM_DIAG

query_fs_info_ops query_path_info_o ps

8B

RAW

NONE

CM_DIAG

8B

RAW

NONE

CM_DIAG

Performance Management Design Guide

30 of 138

3/7/2008

set_path_info_ops query_file_info_op s set_file_info_ops query_disk_info_o ps query_ntap_ext_att r_ops

8B

RAW

NONE

Set Path Information operations (SMB Code = 0x32SubCode = 0x06) Query File Information operations (SMB Code = 0x32SubCode = 0x07) Set File Information operations (SMB Code = 0x32SubCode = 0x08) Query Disk Info operations (SMB Code = 0x80) Query NTAP Extended Attributes Operations (SMB Code = 0x32SubCode = 0x05Info = 0x04) Set NTAP Extended Attributes Operations (SMB Code = 0x32SubCode = 0x06Info = 0x02) Read operations (SMB Code = 0x0A) ReadAndX operations (SMB Code = 0x2E) Read Raw operations (SMB Code = 0x1A) Write operations (SMB Code = 0x0B) WriteAndX operations (SMB Code = 0x2F) Write Raw operations (SMB Code = 0x1D) Queued Write Raw operations (SMB Code = 0x1D) Flush operations (SMB Code = 0x05) Open operations (SMB Code = 0x02) Create operations (SMB Code = 0x03)

CM_DIAG

8B

RAW

NONE

CM_DIAG

8B 8B

RAW RAW

NONE NONE

CM_DIAG CM_DIAG

8B

RAW

NONE

CM_DIAG

set_ntap_ext_attr_o 8B ps read_ops readx_ops read_raw_ops write_ops writex_ops write_raw_ops 8B 8B 8B 8B 8B 8B

RAW

NONE

CM_DIAG

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

queued_write_raw_ 8B ops flush_ops open_ops create_ops 8B 8B 8B

Performance Management Design Guide

31 of 138

3/7/2008

close_ops

8B

RAW

NONE

Close operations (SMB Code = 0x04) Create with extended attributes operations (SMB Code = 0x32SubCode = 0x00) OpenAndX operations (SMB Code = 0x2D) NTCreateAndX operations (SMB Code = 0xA2) NTTransactCreate operations (SMB Code = 0x25SubCode = 0x01)

CM_DIAG

open_ext_ops

8B

RAW

NONE

CM_DIAG

openx_ops nt_create_ops nt_trans_create_op s create_dir_ops delete_dir_ops check_dir_ops delete_ops rename_ops nt_rename_ops seek_ops transact_ops find_first_ops

8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

Create Directory operations CM_DIAG (SMB Code = 0x00) Delete Directory operations CM_DIAG (SMB Code = 0x01) Check Directory operations (SMB Code = 0x10) Delete operations (SMB Code = 0x06) Rename operations (SMB Code = 0x07) NT Rename operations (SMB Code = 0xA5) Seek operations (SMB Code = 0x12) Transact operations (SMB Code = 0x25) Begin search for file operations (SMB Code = 0x32SubCode = 0x01) Resume search for file operations (SMB Code = 0x32SubCode = 0x02) Create Directory with extended attributes operations (SMB Code = 0x32SubCode = 0x0D) Search operations (SMB CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

find_next_ops

8B

RAW

NONE

CM_DIAG

create_dir_ext_ops search_ops

8B 8B

RAW RAW

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

32 of 138

3/7/2008

Code = 0x81) find_close_ops nt_trans_notify_op s lock_byte_range_o ps unlock_byte_range _ops lockx_ops lock_read_ops write_unlock_ops negotiate_ops sess_setup_ops sess_logoff_ops set_sec_ops 8B 8B RAW RAW NONE NONE FindClose2 operations (SMB Code = 0x34) Start directory watch operations (SMB Code = 0x25SubCode = 0x04) Lock Byte Range operations (SMB Code = 0x0C) Unlock Byte Range operations (SMB Code = 0x0D) LockingAndX operations (SMB Code = 0x24) Lock and Read operations (SMB Code = 0x13) Write and Unlock operations (SMB Code = 0x14) Negotiate operations (SMB Code = 0x72) Session Setup operations (SMB Code = 0x73) Session Logoff operations (SMB Code = 0x74) Set Security Descriptor operations (SMB Code = 0x25SubCode = 0x03) Query Security Descriptor operations (SMB Code = 0x25SubCode = 0x06) Unrecognized SMB command code Non supported SMB operations Total number of SMB operations since filer was started CM_DIAG CM_DIAG

8B

RAW

NONE

CM_DIAG

8B 8B 8B 8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

query_sec_ops reject_ops no_support_ops total_ops dfs_refer_ops

8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Get DFS referral operations CM_DIAG (SMB Code = 0x32SubCode

Performance Management Design Guide

33 of 138

3/7/2008

= 0x10) dfs_report_ops echo_ops tree_conn_ops tree_disc_ops ioctl_ops cancel_ops 8B 8B 8B 8B 8B 8B RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE NONE NONE Report DFS inconsistency operations (SMB Code = 0x32SubCode = 0x11) Echo operations (SMB Code = 0x2B) Tree Connect operations (SMB Code = 0x70) Tree Disconnect operations (SMB Code = 0x71) Device IOCTL operations (SMB Code = 0x25SubCode = 0x02) Cancel operations (SMB Code = 0xA4) CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

cifs_session_signing object Counter Name conn_time Size Type Display Unit 8B RAW MSECS Description Total time of a connection to the filer in milliseconds Total time spent calculating security signatures for incoming CIFS requests in milliseconds Priority CM_DIAG Depends on

time_in

8B RAW MSECS

CM_DIAG

time_out

8B RAW MSECS

Total time spent calculating security signatures for outgoing CM_DIAG CIFS requests in milliseconds

cifs_stats object Counter Name curr_sess_cnt max_sess_cnt multi_user_sess_cn Size 4B 4B 8B Type RAW RAW RAW Display Unit NONE NONE NONE Description Number of current sessions High water mark for number of sessions Number of sessions 34 of 138 Priority CM_DIAG CM_DIAG CM_DIAG 3/7/2008 Depends on

Performance Management Design Guide

t sig_sess_cnt client_disc_sess_cn t filer_disc_sess_cnt 8B 8B RAW RAW NONE NONE

with more than one user Number of sessions with signature signing Number of session terminations initiated by client side Number of session terminations initiated by filer side Number of session terminations initiated by both client and filer High water mark for number of credentials attached to a session High water mark for number of trees attached to a session High water mark for number of simultaneous messages attached to a session Number of currently connected users on the filer Number of logon on the filer Number of times a 'null' or 'blank' user was successfully mapped Number of times a new hash table for UIDs is allocated Number of current shares High water mark for number of shares Number of current trees High water mark for number of trees CM_DIAG CM_DIAG

8B

RAW

NONE

CM_DIAG

dup_disc_sess_cnt

8B

RAW

NONE

CM_DIAG

max_cred_sess_cnt

4B

RAW

NONE

CM_DIAG

max_tree_sess_cnt

4B

RAW

NONE

CM_DIAG

max_msg_sess_cnt

4B

RAW

NONE

CM_DIAG

curr_conn_user_cnt 4B logon_cnt map_null_user_cnt 8B 8B

RAW RAW RAW

NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

uid_hash_alloc_cnt curr_share_cnt max_share_cnt curr_tree_cnt max_tree_cnt

8B 4B 4B 4B 8B

RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

35 of 138

3/7/2008

max_fid_tree_cnt max_search_tree_c nt max_core_search_t ree_cnt tid_hash_alloc_cnt curr_open_file_cnt max_open_file_cnt curr_open_dir_cnt max_open_dir_cnt curr_watch_dir_cnt

8B

RAW

NONE

High water mark for number of FIDs attached to one tree High water mark for number of searches attached to one tree High water mark for number of core searches attached to one tree Number of times a new hash table for TIDs is allocated Number of currently open files and directories High water mark for number of open files and directories Number of currently open directories High water mark for number of open directories Number of currently watched directories High water mark for number of watched directories Number of times a new hash table for FIDs is allocated Number of times an attempt is made to fold a file with the version in a snapshot

CM_DIAG

8B

RAW

NONE

CM_DIAG

8B

RAW

NONE

CM_DIAG

8B 4B 4B 4B 4B 4B

RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

max_watch_dir_cnt 4B

fid_hash_alloc_cnt

8B

RAW

NONE

CM_DIAG

fold_attempt_cnt

8B

RAW

NONE

CM_DIAG

fold_rename_cnt

8B

RAW

NONE

Number of times an entry in the queue of files CM_DIAG awaiting folding has to be renamed Number of times an attempt to find a rename match on the queue of CM_DIAG

fold_rename_failur e_cnt

8B

RAW

NONE

Performance Management Design Guide

36 of 138

3/7/2008

files awaiting folding fails Number of times an entry can't be added to the queue of files awaiting folding due to its length limit Number of times when an entry can't be added to the queue of files awaiting folding due to a duplicate Number of times the maximum limit of WAFL concurrent folds has been reached Number of currently allocated locks High water mark for the number of allocated locks Number of OpLock Break from exclusive or batch to level 2 Number of OpLock Break from exclusive or batch to none Number of OpLock Break from level 2 to none Number of OpLock Break ACK before timeout Number of OpLock Break ACK before timeout from Win95 clients Number of OpLock Break ACK before timeout from WinNT clients

fold_overflow_cnt

8B

RAW

NONE

CM_DIAG

fold_duplicate_cnt

8B

RAW

NONE

CM_DIAG

fold_wafl_too_bus y_cnt curr_lock_cnt max_lock_cnt

8B

RAW

NONE

CM_DIAG

4B 4B

RAW RAW

NONE NONE

CM_DIAG CM_DIAG

x_or_batch_to_l2_c 8B nt x_or_batch_to_non e_cnt l2_to_none_cnt 8B

RAW

NONE

CM_DIAG

RAW

NONE

CM_DIAG

8B

RAW

NONE

CM_DIAG

no_break_ack_cnt

8B

RAW

NONE

CM_DIAG

no_break_ack_95_ cnt

8B

RAW

NONE

CM_DIAG

no_break_ack_nt_c nt

8B

RAW

NONE

CM_DIAG

Performance Management Design Guide

37 of 138

3/7/2008

ignored_ack_cnt

8B

RAW

NONE

Number of OpLock Break ACK ignored (e.g. late) Number of OpLock Break which must be delayed Number of authentication requests to Domain Controllers Number of currently active credentials High water mark for number of allocated credentials The most group SIDs found on one credential

CM_DIAG

delayed_break_cnt

8B

RAW

NONE

CM_DIAG

pdc_auth_cnt curr_cred_cnt max_cred_cnt max_sid_cred_cnt built_lgrp_cnt user_lgrp_cnt sid_lgrp_cnt curr_mem_ctrl_blk _cnt curr_mem_ctrl_blk _reserve_cnt max_mem_ctrl_blk _cnt max_mem_ctrl_blk _reserve_cnt

8B 4B 4B 4B 4B 4B 4B 4B 4B

RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Number of built-in local CM_DIAG groups Number of user-defined local groups Number of defined SIDs for local groups Number of current memory control blocks CM_DIAG CM_DIAG CM_DIAG

Number of current memory control blocks in CM_DIAG the reserve pool High water mark for number of memory control blocks High water mark for number of memory control blocks in the reserve pool Number of times a request for memory control block can not be granted Number of times a request for memory control block can not be CM_DIAG

4B

RAW

NONE

4B

RAW

NONE

CM_DIAG

exhaust_mem_ctrl_ blk_cnt exhaust_mem_ctrl_ blk_reserve_cnt

4B

RAW

NONE

CM_DIAG

4B

RAW

NONE

CM_DIAG

Performance Management Design Guide

38 of 138

3/7/2008

granted from the reserve pool wait_mem_ctrl_blk _cnt 8B RAW NONE Number of times waiting for the memory control block to be allocated Number of times waiting for the memory buffer to be allocated Current number of queued authentication requests Current number of queued blocking worker threads Current number of queued timer worker threads Current number of auditing log worker threads Current number of queued SMB RPC worker threads Current number of queued VSCAN worker threads High water mark for number of queued authentication requests High water mark for number of blocking worker threads High water mark for number of timer worker threads High water mark for number of auditing log worker threads High water mark for number of SMB RPC CM_DIAG

wait_mem_buf_cnt

8B

RAW

NONE

CM_DIAG

auth_qlength

4B

RAW

NONE

CM_DIAG

block_qlength

4B

RAW

NONE

CM_DIAG

timer_qlength

4B

RAW

NONE

CM_DIAG

alf_qlength

4B

RAW

NONE

CM_DIAG

rpc_qlength

4B

RAW

NONE

CM_DIAG

offload_qlength

4B

RAW

NONE

CM_DIAG

max_auth_qlength

4B

RAW

NONE

CM_DIAG

max_block_qlength 4B

RAW

NONE

CM_DIAG

max_timer_qlength

4B

RAW

NONE

CM_DIAG

max_alf_qlength max_rpc_qlength

4B 4B

RAW RAW

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

39 of 138

3/7/2008

worker threads max_offload_qleng th copy_align_cnt small_buffer_align _cnt 4B RAW NONE High water mark for number of VSCAN worker threads Count of times a buffer is copied for header alignment Count of times a small buffer is used for header alignment Count of times a large buffer is used for header alignment CM_DIAG

8B

RAW

NONE

CM_DIAG

8B

RAW

NONE

CM_DIAG

large_buffer_align_ 8B cnt read_pipe_busy_err 8B or_cnt write_pipe_busy_er 8B ror_cnt trans_pipe_busy_er ror_cnt read_pipe_broken_ error_cnt write_pipe_broken _error_cnt 8B

RAW

NONE

CM_DIAG

RAW

NONE

Count of read errors due to the 'busy pipe' CM_DIAG condition Count of write errors due to the 'busy pipe' condition Count of transaction errors due to the 'busy pipe' condition CM_DIAG

RAW

NONE

RAW

NONE

CM_DIAG

8B

RAW

NONE

Count of read errors due CM_DIAG to the 'broken pipe' condition Count of write errors due to the 'broken pipe' condition Count of transaction errors due to the 'broken pipe' condition CM_DIAG

8B

RAW

NONE

trans_pipe_broken_ 8B error_cnt

RAW

NONE

CM_DIAG

cpdump object Counter Name cpd_cpcount cpd_why Size 4B 4B Type RAW RAW Display Unit NONE NONE Description Free running CP counter since boot up CP reason CM_PRIV_DIAG_ZAPI| CM_CP_DUMP CM_PRIV_DIAG_ZAPI| 40 of 138 3/7/2008 Priority

Performance Management Design Guide

CM_CP_DUMP cpd_alloc_loo ps cpd_dirty_all oc_blks cpd_inofile_b lks 4B 4B 4B RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE iterations to alloc write_allocation files dirty blks write allocated in CP inofile blks write allocated in CP inodes cleaned in CP read messages returned in CP dirty buf suspend water mark dirty buf suspend replay watermark dirty buf high watermark dirty inodes high water mark available buffers low watermark available buffers high watermark available vbufs low watermark avail vbufs high watermark vbufs watermark to trigger a CP wafl_load_buf misses wafl_load_buf buffers loaded wafl_read_buf read buffers active file blocks read CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

cpd_inodes_cl 4B eaned cpd_reads_ret urned 4B

cpd_water_dir 4B ty_buf_susp cpd_water_dir ty_buf_susp_re 4B play cpd_water_dir 4B ty_buf_high cpd_water_dir 4B ty_ino_high cpd_water_av ail_buf_low cpd_water_av ail_buf_high cpd_water_av ail_vbuf_low cpd_water_av ail_vbuf_high cpd_water_tri gger_cp_vbufs cpd_io_buf_ miss_cnt cpd_io_buf_l oad_cnt 4B 4B 4B 4B 4B 4B 4B

cpd_io_blks_r 4B ead cpd_io_blks0 _active cpd_io_blks0 4B 4B

snapmap file blocks CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

Performance Management Design Guide

41 of 138

3/7/2008

_snapmap cpd_io_blks0 _summary cpd_io_blks0 _spacemap cpd_io_blks0 _kireeti cpd_io_blks0 _container cpd_io_blks0 _inofile cpd_io_blks0 _metadata cpd_io_blks0 _regular cpd_io_blks0 _snapcnt cpd_io_blks1 _active cpd_io_blks1 _snapmap cpd_io_blks1 _summary cpd_io_blks1 _spacemap cpd_io_blks1 _kireeti cpd_io_blks1 _container cpd_io_blks1 _inofile cpd_io_blks1 _metadata cpd_io_blks1 _regular cpd_io_blks1 _snapcnt 4B 4B 4B 4B 4B 4B 4B RAW RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE NONE NONE NONE

read summary file blocks read spacemap file blocks read kireeti file blocks read container file blocks read inofile file blocks read CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

metadata file blocks CM_PRIV_DIAG_ZAPI|CM_CP_DUMP read regular file blocks read count of snapmap blks read summed over all the snapshots active file blocks read CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

4B

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

4B 4B 4B 4B 4B 4B 4B 4B 4B 4B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

snapmap file blocks CM_PRIV_DIAG_ZAPI|CM_CP_DUMP read summary file blocks read spacemap file blocks read kireeti file blocks read container file blocks read inofile file blocks read CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

metadata file blocks CM_PRIV_DIAG_ZAPI|CM_CP_DUMP read regular file blocks read count of snapmap blks read summed CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

Performance Management Design Guide

42 of 138

3/7/2008

over all the snapshots cpd_ra_reque sted cpd_ra_read 4B 4B RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE read-ahead blocks requested r-a blocks actually read r-a blocks actually resident CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

cpd_ra_incore 4B cpd_ra_past_e 4B of cpd_ra_holes cpd_ra_noblk s cpd_ra_noms gs cpd_ra_nobuf s 4B 4B 4B 4B

r-a blocks requested CM_PRIV_DIAG_ZAPI|CM_CP_DUMP past eof r-a blocks requested CM_PRIV_DIAG_ZAPI|CM_CP_DUMP from holes r-a blocks requested CM_PRIV_DIAG_ZAPI|CM_CP_DUMP from lev-0 r-a blocks dropped for lack of msgs r-a blocks dropped for lack of bufs estimated number of dirty bufs no. of dirty bufs the CP begins with no. of dirty inofile bufs the CP begins with no. of bufs the CP knows it will dirty as of CP start CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

cpd_dbufs_tot 4B al cpd_dbufs_in coming cpd_dbufs_in coming_inofile 4B 4B

cpd_dbufs_in coming_commi 4B tment cpd_dbufs_in ofile cpd_dbufs_all ocfile cpd_dbufs_in direct cpd_dbufs_nd x0 4B

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

RAW

NONE

no. of blocks dirtied from inofile flush in CM_PRIV_DIAG_ZAPI|CM_CP_DUMP current CP no. of blocks dirtied from allocfile flush CM_PRIV_DIAG_ZAPI|CM_CP_DUMP in current CP no. of blocks dirtied from indirect blocks CM_PRIV_DIAG_ZAPI|CM_CP_DUMP in current CP no. of blocks dirtied from ndx0 file CM_PRIV_DIAG_ZAPI|CM_CP_DUMP blocks in current CP

4B

RAW

NONE

4B

RAW

NONE

4B

RAW

NONE

Performance Management Design Guide

43 of 138

3/7/2008

cpd_dbufs_w alloc cpd_dbufs_wr itten cpd_dbufs_co mmitted cpd_time_star t cpd_time_end cpd_time_rea d cpd_time_nex t_log_nearly_f ull cpd_time_nvl og_full_susp cpd_time_dirt y_buf_susp cpd_cycles_w afl_threads_idl e cpd_cycles_w afl_threads_sle eping

4B 4B 4B 4B 4B 4B 4B

RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE MSECS MSECS MSECS

no. of dbufs write_allocated so far no. of dbufs sent to raid so far no. of dbufs committed to disk so far start time of CP in msecs end time of CP in msecs total time for CP reads in msecs

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

time in msecs when MSECS the next nvlog was nearly full time in msecs when MSECS clients suspend for full nvlog time in msecs when MSECS clients suspend for too many dirty bufs NONE cpu cycles that wafl threads were idle cpu cycles that wafl threads were sleeping cpu cycles that wafl threads were runnable cpu cycles that wafl threads were waiting for mutex cpu cycles that wafl threads were runnable while holding the mutex cpu cycles that wafl threads were running

4B

RAW

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

4B

RAW

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

8B

RAW

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

8B

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

cpd_cycles_w afl_threads_run 8B nable cpd_cycles_w afl_threads_mu tex_wait cpd_cycles_w afl_threads_mu tex_runnable 8B

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

8B

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

cpd_cycles_w 8B afl_threads_run

RAW

NONE

CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

Performance Management Design Guide

44 of 138

3/7/2008

ning cpd_cycles_w afl_threads_pre empted cpd_phase_ti me cpd_total_tim e 8B 4B 4B RAW RAW RAW NONE MSECS MSECS cpu cycles that wafl threads were preempted time taken in CP phases total time taken by CP in msecs CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP CM_PRIV_DIAG_ZAPI|CM_CP_DUMP

cpdump_rvol object Counter Name Size Type Display Description Unit Priority

cpd_rvol_cpcount

Free running CP 4B RAW NONE counter since boot up number of hot disks 4B RAW NONE avoided during write-alloc number of degraded disks 4B RAW NONE avoided during write-alloc time taken for CP 4B RAW MSECS setup stage in msecs per vol time taken for CP pre phase0 4B RAW MSECS stage in msecs per vol 4B RAW MSECS time taken

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_raidvol_hot_disks

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_raidvol_degraded_disks

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_setup

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_pre_p0

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p0_snap_del

CM_PRIV_DIAG_ZAPI|CM_CP_DU

Performance Management Design Guide

45 of 138

3/7/2008

for CP phase0 snap delete stage in msecs per vol time taken for CP phase1 4B RAW MSECS clean normal files per vol time taken for CP 4B RAW MSECS phase1 clean quota files per vol

cpd_phases_vol_p1_clean

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p1_quota

CM_PRIV_DIAG_ZAPI|CM_CP_DU

time taken for CP phase2 cpd_phases_vol_p2v_refcount 4B RAW MSECS volume refcount in msecs per vol time taken for CP phase2 4B RAW MSECS volume inofile in msecs per vol time taken for CP phase2 4B RAW MSECS volume inode stage in msecs per vol time taken for CP phase2 4B RAW MSECS volume bitmap stage in msecs per

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2v_inofile

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2v_ino

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2v_bm

CM_PRIV_DIAG_ZAPI|CM_CP_DU

Performance Management Design Guide

46 of 138

3/7/2008

vol time taken for CP phase2 4B RAW MSECS volume snapmap in msecs per vol time taken for CP phase2 4B RAW MSECS volume volinfo in msecs per vol time taken for CP phase2 4B RAW MSECS volume container map stage per vol

cpd_phases_vol_p2v_snap

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2v_volinfo

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2v_cont

CM_PRIV_DIAG_ZAPI|CM_CP_DU

time taken for CP phase2 aggr cpd_phases_vol_p2a_refcount 4B RAW MSECS refcount stage in msecs per vol

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2a_inofile

time taken for CP phase2 aggr 4B RAW MSECS CM_PRIV_DIAG_ZAPI|CM_CP_DU inofile stage in msecs per vol time taken for CP phase2 aggr 4B RAW MSECS inode stage in msecs per vol 4B RAW MSECS time taken for CP

cpd_phases_vol_p2a_ino

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2a_bm

CM_PRIV_DIAG_ZAPI|CM_CP_DU

Performance Management Design Guide

47 of 138

3/7/2008

phase2 aggr bitmap stage in msecs per vol time taken for CP phase2 aggr 4B RAW MSECS snapmap stage in msecs per vol time taken for CP phase2 aggr 4B RAW MSECS volinfo stage in msecs per vol time taken for CP phase2 4B RAW MSECS flush stage in msecs per vol time taken for CP phase2 4B RAW MSECS finish stage in msecs per vol time taken for CP phase3 wait 4B RAW MSECS stage in msecs per vol time taken for CP phase3 4B RAW MSECS volume volinfo in msecs per vol

cpd_phases_vol_p2a_snap

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2a_volinfo

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2_flush

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p2_finish

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p3_wait

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p3v_volinfo

CM_PRIV_DIAG_ZAPI|CM_CP_DU

Performance Management Design Guide

48 of 138

3/7/2008

cpd_phases_vol_p3a_volinfo

time taken for CP phase3 aggr 4B RAW MSECS volinfo stage in msecs per vol time taken for CP phase3 4B RAW MSECS finish stage in msecs per vol time taken for CP phase4 4B RAW MSECS finish stage in msecs per vol

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p3_finish

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_p4_finish

CM_PRIV_DIAG_ZAPI|CM_CP_DU

cpd_phases_vol_total

total time taken by CP 4B RAW MSECS CM_PRIV_DIAG_ZAPI|CM_CP_DU in msecs per volume

cpdump_rg object Counter Name cpd_rg_local_id Size Type Display Description Unit Priority

Local id 4B RAW NONE used for this CM_PRIV_DIAG_ZAPI|CM_CP_DUM raid group AA cache count for 4B RAW NONE the raidgroup AA mean 4B RAW NONE for the raidgroup max AA 4B RAW NONE for the raidgroup

cpd_rg_aa_cache_count

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

cpd_rg_aa_rg_mean

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

cpd_rg_aa_rtg_maxaa

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

Performance Management Design Guide

49 of 138

3/7/2008

cpd_rg_wcp_ntetrises

number of tetrises 4B RAW NONE written in this CP number of blocks 4B RAW NONE written in this CP total of 4B RAW MSECS completion times max of 4B RAW MSECS completion times

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

cpd_rg_wcp_nblocks

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

cpd_rg_wcp_total_ms

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

cpd_rg_wcp_max_ms

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

length of the tetris cpd_rg_wcp_max_ms_nblks 4B RAW MSECS that took max time

CM_PRIV_DIAG_ZAPI|CM_CP_DUM

cpdump_disk object Counter Name Size Type Display Description Unit Priority

cpd_disk_rg

raid group id to which 4B RAW NONE the disk belongs msgs by cleaner for 4B RAW NONE read of a disk blks read 4B RAW NONE by cleaner for a disk total of completion 4B RAW MSECS times per disk by cleaner 4B RAW MSECS max time taken per

CM_PRIV_DIAG_ZAPI|CM_CP_D

cpd_disk_walloc_msgs

CM_PRIV_DIAG_ZAPI|CM_CP_D

cpd_disk_walloc_blks

CM_PRIV_DIAG_ZAPI|CM_CP_D

cpd_disk_walloc_total_ms

CM_PRIV_DIAG_ZAPI|CM_CP_D

cpd_vol_walloc_max_ms

CM_PRIV_DIAG_ZAPI|CM_CP_D

Performance Management Design Guide

50 of 138

3/7/2008

disk by cleaner

cpd_disk_during_CP_msgs

msgs by others 4B RAW NONE during CP CM_PRIV_DIAG_ZAPI|CM_CP_D for read of a disk blks read by others 4B RAW NONE during CP for a disk

cpd_disk_during_CP_blks

CM_PRIV_DIAG_ZAPI|CM_CP_D

total of completion times per cpd_disk_during_CP_total_ms 4B RAW MSECS disk by others during CP max time taken per 4B RAW MSECS disk by others during CP

CM_PRIV_DIAG_ZAPI|CM_CP_D

cpd_vol_during_CP_max_ms

CM_PRIV_DIAG_ZAPI|CM_CP_D

cpdump_fvol object Counter Name cpd_fvol_aa_cache _count cpd_fvol_aa_rg_me an cpd_fvol_aa_rtg_m axaa Size 4B 4B 4B Type RAW RAW RAW Display Unit NONE NONE NONE Description AA cache count for the raidgroup of flexvol AA mean for the raidgroup of flexvol max AA for the raidgroup of flexvol Priority CM_PRIV_DIAG_ZAPI| CM_CP_DUMP CM_PRIV_DIAG_ZAPI| CM_CP_DUMP CM_PRIV_DIAG_ZAPI| CM_CP_DUMP

Depends

Performance Management Design Guide

51 of 138

3/7/2008

processor object Counter Name processor_busy processor_elapse d_time sk_switches hard_switches domain_busy Size 8B Type PERCT DELTA NODIS P RATE RATE PERCT Display Unit PERCT DELTA NODISP PER_SEC PER_SEC PERCT Description Percentage of elapsed time that the processor is executing non-idle processes Wall-clock time since boot used for calculating processor utilization Number of sk switches per second Number of context switches per second Array of processor time in percentage spent in various domains Priority CM_BASIC

Depe

processor_elapsed_

8B 4B 4B 8B

CM_BASIC CM_DIAG CM_DIAG CM_DIAG|C M_ASUP_HO URLY

processor_elapsed_

sis object Counter Name Size Type RAW RAW RAW RAW RAW RAW RAW RAW RAW Display Unit NONE NONE NONE NONE NONE NONE NONE NONE NONE Description Number of succeeded operations since reboot Number of deferred operations since reboot Priority CM_DIAG CM_DIAG Depends on

num_succeeded_ 4B op num_deferred_o p 4B

num_stopped_op 4B num_failed_op total_blk_cnt total_op_time blks_used blks_shared share_saved_blk 4B 8B 4B 4B 4B 4B

Number of stopped operations CM_DIAG since reboot Number of failed operations since reboot Total amount of blocks processed in the SIS volume Total operation time of the SIS volume in seconds Number of blocks used in the SIS volume Number of shared blocks in the SIS volume Number of blocks saved by CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

52 of 138

3/7/2008

s num_sis_files max_refcount num_blks_mism atch ref_count_histog ram 4B 4B 4B 4B RAW RAW RAW RAW NONE NONE NONE NONE

sharing in the SIS volume Number of dense files in the SIS volume Max refcount value on each block Number of blocks which had same fingerprint but failed bcmp check CM_DIAG CM_DIAG CM_DIAG

Histogram of reference counts CM_DIAG in the SIS volume

disk object Counter Name display_name raid_name raid_type disk_speed disk_capacity total_transfers Size 64B 64B 8B 8B 4B 8B Type Display Unit Description Name of the disk Name of the disk in raid terminology Raid type of this disk Disk RPM Disk capacity in MB Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC

Depends on

STRING NONE STRING NONE STRING NONE STRING NONE RAW RATE AVERA GE MB

Total number of disk PER_S operations involving data EC transfer initiated per second Average number of blocks NONE transferred in each user read operation Number of disk read operations initiated each PER_S second for retrieving data or EC metadata associated with user requests Average number of blocks NONE transferred in each user write operation Number of disk write PER_S operations initiated each EC second for storing data or metadata associated with user

user_read_chain

8B

CM_BASIC

user_reads

user_reads

8B

RATE

CM_BASIC

user_write_chai n

8B

AVERA GE

CM_BASIC

user_writes

user_writes

8B

RATE

CM_BASIC

Performance Management Design Guide

53 of 138

3/7/2008

requests user_writes_in_s kip_mask 8B RATE Number of disk write IOs that PER_S were executed as part of a CM_BASIC EC skip-mask write Number of disk skip-write operations initiated each PER_S second for storing data or EC metadata associated with user requests Average number of blocks transferred in each consistency NONE point read operation during a CP Number of disk read PER_S operations initiated each EC second for consistency point processing Average number of blocks NONE transferred in each guaranteed read operation Number of disk read PER_S operations initiated each EC second for raid reconstruct or scrubbing activities Average number of blocks NONE transferred in each guaranteed write operation Number of write read PER_S operations initiated each EC second for raid reconstruct or scrubbing activities Average latency per block in USECS microseconds for user read operations Number of blocks transferred PER_S for user read operations per EC second Average latency per block in USECS microseconds for user write operations PER_S Number of blocks trasnferred

user_skip_write _ios

8B

RATE

CM_BASIC

cp_read_chain

8B

AVERA GE

CM_BASIC

cp_reads

cp_reads

8B

RATE

CM_BASIC

guarenteed_read _chain guarenteed_read s guarenteed_writ e_chain guarenteed_writ es user_read_latenc y user_read_block s user_write_laten cy user_write_bloc

8B

AVERA GE

CM_BASIC

guarenteed_reads

8B

RATE

CM_BASIC

8B

AVERA GE

CM_BASIC

guarenteed_writes

8B

RATE

CM_BASIC

8B

AVERA GE RATE AVERA GE RATE

CM_BASIC

user_read_blocks

8B

CM_BASIC

8B 8B

CM_BASIC CM_BASIC

user_write_blocks

Performance Management Design Guide

54 of 138

3/7/2008

ks

EC

for user write operations per second CM_BASIC

skip_blocks

8B

RATE AVERA GE RATE AVERA GE RATE AVERA GE RATE

Number of blocks skipped in PER_S skip-mask write operations per EC second Average latency per block in USECS microseconds for consistency point read operations Number of blocks transferred PER_S for consistency point read EC operations per second Average latency per block in USECS microseconds for guaranteed read operations Number of blocks transferred PER_S for guaranteed read operations EC per second Average latency per block in USECS microseconds for guaranteed write operations

cp_read_latency

8B

CM_BASIC

cp_read_blocks

cp_read_blocks guarenteed_read _latency guarenteed_read _blocks guarenteed_writ e_latency guarenteed_writ e_blocks disk_busy base_for_disk_b usy

8B

CM_BASIC

8B

CM_BASIC

guarenteed_read_bloc

8B

CM_BASIC

8B

CM_BASIC

guarenteed_write_blo

8B

Number of blocks transferred PER_S for guaranteed write operations CM_BASIC EC per second Percentage of time there was PERCT at least one outstanding request to the disk CM_BASIC base_for_disk_busy

8B

PERCT

8B

DELTA DELTA Time base for disk_busy NODIS NODISP calculation P AVERA GE AVERA GE DELTA Average number of IOs issued to the diskfor which we NONE have not yet received the response NONE NONE Number of IOs queued to the disk but not yet issued Disk I/O latency histogram

CM_BASIC

io_pending

8B

CM_DIAG

base_for_disk_busy

io_queued

8B

CM_DIAG CM_DIAG

base_for_disk_busy

disk_io_latency_ 8B histogram

Performance Management Design Guide

55 of 138

3/7/2008

dump object Counter Name Size p1-ino p1-dir p1-str-ino p1-str-dir p1-acl p3-dir p3-write p4-ino p4-write INO-blks ino ino-blks ra-file ra-p45 tape p3-dirs 8B 8B 8B 8B 8B 8B 8B 8B 8B 4B 4B 4B 4B 4B 4B 8B Type RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAT E Display Unit NONE NONE NONE Description Number of regular inodes marked to be dumped in Phase 1 Number of directories marked to be dumped in Phase 1 NT stream inodes marked to be dumped in Phase 1 Priority CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

Number of NT stream NONE directories marked to be dumped in Phase 1 NONE NONE NONE NONE NONE Number of acl inodes marked to be dumped in Phase 1 Number of directories dumped in Phase 3 Total amount of data written (KB) to tape in Phase 3 Number of regular inodes dumped in Phase 3 Total amount of data written (KB) to tape in Phase 4

1 if dump is in the state of NONE reading the INO blocks; 0 otherwise NONE 1 if dump is in the state of reading an inode; 0 otherwise

1 if dump is in the state of NONE reading blocks for an inode; 0 otherwise NONE 1 if dump is in the state of prefetching a file; 0 otherwise

1 if dump is in the state of NONE prefetching in Phases 4 or 5; 0 otherwise NONE 1 if dump is in the state of writing to tape; 0 otherwise

PER_S Directories dumped in Phase 3 EC (inodes per sec)

Performance Management Design Guide

56 of 138

3/7/2008

p3-read

RAT 8B E 8B RAT E

KB_P Directory data read from WAFL ER_SE CM_DIAG (KB per sec) C KB_P Directory data written to tape ER_SE (KB per sec) C Number of directories that have PER_S completed readahead (inodes per EC sec) PER_S Number of Phase 3 level 0 EC readaheads (per sec) PER_S Number of Phase 3 level 0 EC readahead blocks (per sec) PER_S Inodes dumped (per sec) EC KB_P Inode data read from WAFL ER_SE (KB per sec) C KB_P Inode data written to tape (KB ER_SE per sec) C PER_S Number of inodes that have EC completed readahead (per sec) PER_S Number of Phase 4 level 0 EC readaheads (per sec) PER_S Number of Phase 4 level 0 EC readahead blocks (per sec) CM_DIAG

p3-write

p3-ra-inos p3-ra-L0 p3-ra-L0-blks p4-inos p4-read

RAT 8B E 8B 8B 8B 8B RAT E RAT E RAT E RAT E

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

p4-write p4-ra-inos p4-ra-L0 p4-ra-L0-blks

RAT 8B E 8B 8B 8B RAT E RAT E RAT E

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

ext_cache object Counter Name state state-string Size 4B 16B Type RAW STRING RAW RAW Display Unit NONE NONE NONE NONE Description State of external cache State of external cachetextual representation Number of cache objects available Number of aggregates or traditional volumes using Priority CM_BASIC CM_BASIC CM_BASIC CM_DIAG Depends on

cache_objects 4B pvols 4B

Performance Management Design Guide

57 of 138

3/7/2008

the cache doneq_loads doneq_flushe s doneq_bufs 4B RAW Number of times external NONE cache completion queue was CM_DIAG reloaded when empty Number of times external NONE cache completion queue was CM_DIAG flushed after timeout Number of buffers NONE currently in cache completion queue Maximum number of I/O NONE ever concurrently in progress NONE NONE Total uncorrectable ECC errors reported Set if block checksums are enabled CM_DIAG CM_ADVAN CED CM_DIAG CM_BASIC

4B

RAW

4B

RAW

max_qdepth uecc_total block_checks ums

4B 4B 4B

RAW RAW RAW

ext_cache_obj object Counter Name type uptime blocks associativity sets usage accesses_total accesses accesses_sync Size 16B 4B 4B 4B 4B 8B 8B 8B 8B Type STRIN G RAW RAW RAW RAW RAW RAW DELTA DELTA Display Unit NONE NONE NONE NONE NONE PERCT NONE NONE NON_ZERO Description Type of external cache Time in ms since external cache was last enabled Size of the external cachein 4KB blocks Cache associativity Priority CM_BASIC CM_DIAG CM_BASIC CM_DIAG Depends on

Number of associative sets in CM_DIAG external cache Percentage of blocks in external cache currently containing valid data Total external cache accesses External cache accesses per second External cache accesses CM_BASIC CM_ADVA NCED CM_BASIC CM_DIAG

Performance Management Design Guide

58 of 138

3/7/2008

fast path hit hit_flushq hit_once hit_age hit_normal_lev 0 hit_metadata_fi le hit_directory hit_redirect hit_indirect hit_type hit_partial 8B 8B 8B 8B 8B RATE RATE RATE RATE RATE PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC Number of wafl buffers served off the external cache Number of wafl flushq buffers served off the external cache Number of wafl once buffers served off the external cache Number of wafl age buffers served off the external cache Number of normal level 0 wafl buffers served off the external cache Number of metadata file buffers served off the external cache Number of directory buffers served off the external cache Number of redirected buffers served off the external cache Number of indirect file buffers served off the external cache block types of hits Number of hits where somenot allchained buffers were found in external cache CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_BASIC

8B 8B 8B 8B 8B 8B

RATE RATE RATE RATE DELTA RATE

PER_SEC PER_SEC PER_SEC PER_SEC NON_ZERO PER_SEC

CM_BASIC CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_DIAG

hit_sync

8B

RATE

Number of wafl buffers NON_ZERO served off the external cache fast path Number of wafl flushq NON_ZERO buffers served off the external cache - fast path Number of wafl once buffers NON_ZERO served off the external cache fast path Number of wafl age buffers NON_ZERO served off the external cache fast path NON_ZERO Number of normal level 0

CM_DIAG

hit_flushq_sync 8B

RATE

CM_DIAG

hit_once_sync

8B

RATE

CM_DIAG

hit_age_sync hit_normal_lev

8B 8B

RATE RATE

CM_DIAG CM_DIAG

Performance Management Design Guide

59 of 138

3/7/2008

0_sync hit_metadata_fi le_sync hit_directory_s ync hit_redirect_sy nc miss miss_flushq miss_once miss_age

wafl buffers served off the external cache - fast path 8B RATE Number of metadata file NON_ZERO buffers served off the external cache - fast path Number of directory buffers NON_ZERO served off the external cache fast path Number of redirect buffers NON_ZERO served off the external cache fast path PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC External cache misses External cache misses accessing flushq buffers External cache misses accessing once buffers External cache misses accessing age buffers External cache misses accessing normal level 0 buffers External cache misses accessing metadata file buffers External cache misses accessing directory buffers External cache misses accessing indirect file buffers block types of misses External cache misses - fast path CM_DIAG

8B

RATE

CM_DIAG

8B 8B 8B 8B 8B

RATE RATE RATE RATE RATE RATE

CM_DIAG CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG

miss_normal_le 8B v0 miss_metadata _file miss_directory miss_indirect miss_type miss_sync miss_flushq_sy nc miss_once_syn c miss_age_sync 8B 8B 8B 8B 8B 8B

RATE RATE RATE DELTA RATE RATE

PER_SEC PER_SEC PER_SEC NON_ZERO NON_ZERO

CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG

External cache misses NON_ZERO accessing flushq buffers - fast path External cache misses NON_ZERO accessing once buffers - fast path External cache misses NON_ZERO accessing age buffers - fast path

8B

RATE

CM_DIAG

8B

RATE

CM_DIAG

Performance Management Design Guide

60 of 138

3/7/2008

miss_normal_le 8B v0_sync miss_metadata _file_sync miss_directory _sync lookup_reject 8B

RATE

External cache misses NON_ZERO accessing normal level 0 buffers - fast path External cache misses NON_ZERO accessing metadata file buffers - fast path External cache misses NON_ZERO accessing directory buffers fast path PER_SEC Number of lookups not done because the buffer came from flushq

CM_DIAG

RATE

CM_DIAG

8B

RATE

CM_DIAG

8B

RATE

CM_DIAG

lookup_reject_s 8B ync lookup_reject_ normal_l0 lookup_reject_i o lookup_reject_l ock lookup_chains 8B 8B 8B 8B

RATE

Number of sync lookups not NON_ZERO done because the buffer came from flushq - fast path PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PERCT NON_ZERO PER_SEC Number of lookups not done because the buffer is normal L0

CM_DIAG

RATE RATE RATE RATE RATE RATE PERCT PERCT RATE

CM_DIAG

Number of lookups not done CM_ADVA because no I/O were available NCED Number of lookups not done CM_ADVA because of lock hierarchy NCED Number of chains looked up CM_ADVA in external cache NCED Number of blocks looked up CM_ADVA in chains in external cache NCED Number of lookups not done because lwr context mismatch External cache hit rate External cache hit rate - fast path Number of wafl buffers inserted into the external cache Number of mbuf-backed wafl buffers inserted into the external cache Number of zero-copy mbufbacked wafl buffers inserted into the external cache CM_DIAG CM_BASIC CM_DIAG CM_BASIC accesses accesses_sync

lookup_chain_c 8B nt lookup_lwr_mi smatch hit_percent hit_percent_syn c inserts 8B 8B 8B 8B

inserts_mbuf inserts_mbuf_z erocopy

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

Performance Management Design Guide

61 of 138

3/7/2008

inserts_flushq

8B

RATE

PER_SEC

Number of flushq buffers inserted into the external cache Number of once buffers inserted into the external cache Number of age buffers inserted into the external cache Number of normal level 0 wafl buffers inserted into the external cache Number of metadata file buffers inserted into the external cache Number of directory buffers inserted into the external cache Number of indirect blocks inserted into the external cache Number of buffers not inserted into external cache because they were ineligible

CM_DIAG

inserts_once

8B

RATE

PER_SEC

CM_DIAG

inserts_age inserts_normal _lev0 inserts_metadat a_file inserts_director y inserts_indirect insert_rejects_ misc insert_rejects_c ompressed insert_rejects_p resent insert_rejects_f lushq

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_BASIC

8B

RATE

PER_SEC

CM_BASIC

8B

RATE

PER_SEC

CM_BASIC

8B

RATE

PER_SEC

CM_BASIC

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

Number of buffers not inserted into external cache CM_DIAG because they were compressed Number of buffers not inserted into external cache because they were already in Number of buffers not inserted into external cache because they came from flushq Number of buffers not inserted into external cache because they are normal level0 Number of buffers not inserted into external cache due to resource constraints CM_ADVA NCED

8B

RATE

PER_SEC

8B

RATE

PER_SEC

CM_DIAG

insert_rejects_n ormal_lev0 insert_rejects_t hrottle

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_ADVA NCED

Performance Management Design Guide

62 of 138

3/7/2008

insert_rejects_t hrottle_io insert_rejects_t hrottle_refill insert_rejects_t hrottle_mem insert_rejects_c ache_reuse insert_rejects_v bn_invalid insert_rejects_p vbn_aliased insert_rejects_ mbuf_nobuf

8B

RATE

PER_SEC

Number of buffers not inserted into external cache because no IOs are available Number of buffers not inserted into external cache because no replacement bufs available Number of buffers not inserted into external cache because no vbufs available Number of buffers not inserted into external cache because cache-reuse set on volume Number of buffers not inserted into external cache because VBN is invalid Number of buffers not inserted into external cache because PVBN is aliased Number of mbuf-backed buffers not inserted into external cache because no buffer is available Number of mbuf-backed buffers not inserted into external cache because free mbufs are low

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

8B

RATE

PER_SEC

CM_DIAG

insert_rejects_ mbuf_low reuse_percent evicts

8B

RATE

PER_SEC

CM_DIAG

8B 8B

PERCT RATE

PERCT PER_SEC

Ratio of cache lookup hits to CM_ADVA insertsas a percentage NCED Number of blocks evicted from the external cache to make room for new blocks Number of blocks evicted from the external cache that were used once or more. Number of single block cache read IOs Number of cache read IOs Number of blocks in all cache read IOs CM_BASIC

inserts

evicts_ref readio_solitary readio_chains readio_blocks

8B 8B 8B 8B

RATE RATE RATE RATE

PER_SEC PER_SEC PER_SEC PER_SEC

CM_BASIC CM_BASIC CM_BASIC CM_BASIC

Performance Management Design Guide

63 of 138

3/7/2008

readio_in_fligh t readio_max_in _flight readio_avg_cha inlength

8B 8B 8B

RAW RAW AVERA GE AVERA GE DELTA DELTA RATE RATE RATE RAW RAW AVERA GE AVERA GE DELTA DELTA RAW

NONE NONE NONE MSECS NON_ZERO NON_ZERO PER_SEC PER_SEC PER_SEC NONE NONE NONE MSECS NON_ZERO NON_ZERO NONE

Number of read IOs currently in flight Maximum number of read IOs ever concurrently in flight Average chainlength of read I/Os Average latency of read I/Os Histogram of latency of issued read I/O Histogram of chain lengths of issued read I/O Number of single block cache write IOs Number of cache write IOs Number of blocks in all cache write IOs Number of write IOs currently in flight Maximum number of write IOs ever concurrently in flight Average chainlength of write I/Os Average latency of write I/Os Histogram of latency issued write I/O Histogram of chain lengths of issued write I/O Number of non-referenced blocks currently in the external cache Number of blocks currently in the external cache referenced once Number of blocks currently in the external cache referenced twice Number of blocks currently

CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED CM_ADVA writeio_chains writeio_chains readio_chains readio_chains

readio_avg_late 8B ncy readio_latency 4B

readio_chain_le 4B ngths writeio_solitary writeio_chains writeio_blocks writeio_in_flig ht writeio_max_in _flight writeio_avg_ch ainlength writeio_avg_lat ency writeio_latency writeio_chain_l engths blocks_ref0 8B 8B 8B 8B 8B 8B 8B 4B 4B 8B

blocks_ref1

8B

RAW

NONE

blocks_ref2 blocks_ref3

8B 8B

RAW RAW

NONE NONE

Performance Management Design Guide

64 of 138

3/7/2008

in the external cache referenced thrice blocks_ref4 8B RAW NONE Number of blocks currently in the external cache referenced four times Number of blocks currently in the external cache referenced five times Number of blocks currently in the external cache referenced six times Number of blocks currently in the external cache referenced seven times or more

NCED CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED

blocks_ref5

8B

RAW

NONE

blocks_ref6

8B

RAW

NONE

blocks_ref7 blocks_ref0_arr ivals blocks_ref1_arr ivals blocks_ref2_arr ivals blocks_ref3_arr ivals blocks_ref4_arr ivals blocks_ref5_arr ivals blocks_ref6_arr ivals blocks_ref7_arr ivals lru_ticks invalidates

8B

RAW

NONE

8B 8B 8B 8B 8B 8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RATE

NONE NONE NONE NONE NONE NONE NONE NONE NONE PER_SEC

Number of block transitions CM_ADVA to non-referenced state NCED Number of block transitions CM_ADVA to once referenced state NCED Number of block transitions CM_ADVA to twice referenced state NCED Number of block transitions CM_ADVA to thrice referenced state NCED Number of block transitions CM_ADVA to four times referenced state NCED Number of block transitions CM_ADVA to five times referenced state NCED Number of block transitions CM_ADVA to six times referenced state NCED Number of block transitions CM_ADVA to seven times or more NCED referenced state Number of times LRU clock ticked Number of blocks invalidated in the external cache CM_DIAG CM_BASIC

Performance Management Design Guide

65 of 138

3/7/2008

flexcache object Counter Name per_ops per_ops_total hit_ops_perct miss_ops_perct proxy_ops_perct hit_pure_perct total hit_verify_perct total1 miss_pure_perct total2 miss_verify_perct total3 proxy_perct total4 bandwidth_savings _perct bandwidth_to_clien ts bandwidth_to_origi Size 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type DELTA DELTA PERCT PERCT PERCT PERCT DELTA PERCT DELTA NODISP PERCT DELTA NODISP PERCT DELTA NODISP PERCT DELTA NODISP PERCT RATE RATE Display Unit NONE NONE PERCT PERCT PERCT PERCT NONE PERCT DELTA NODISP PERCT DELTA NODISP PERCT DELTA NODISP PERCT DELTA NODISP PERCT Description Client ops stats on FlexCache volume Total ops on FlexCache volume by operation Per-op hit rate on FlexCache volume Per-op miss rate on FlexCache volume Per-op proxy rate on FlexCache volume Hit rate on FlexCache volume without verify Total ops on FlexCache volume Hit rate on FlexCache volume with verify Total ops on FlexCache volume Miss rate on FlexCache volume without verify Total ops on FlexCache volume Miss rate on FlexCache volume with verify Total ops on FlexCache volume Proxy rate on FlexCache volume Total ops on FlexCache volume Bandwidth savings on FlexCache volume Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG total4 total3 total2 total1 per_ops_total per_ops_total per_ops_total total Depends on

bandwidth_to_cli

Total traffic on B_PER_S FlexCache volume to EC clients B_PER_S Total traffic on

Performance Management Design Guide

66 of 138

3/7/2008

n inode_type_stats delegations delegation_fetches delegation_returns delegation_resets delegation_lookup_ queue_length_hit delegation_lookup_ hit_rate delegation_lookup_ ops delegation_lookup_ queue_length_miss delegation_lookup_ miss_rate delegation_lookup_ ops1 inodes_tossed blocks_tossed trunc_blks_capacit y trunc_blks_usage trunc_blks_relative _capacity trunc_blks_relative _usage trunc_blks_ideal_a vail trunc_blks_ideal_a 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B DELTA RAW DELTA DELTA DELTA DELTA PERCT DELTA DELTA PERCT DELTA NODISP DELTA DELTA RAW RAW RAW RAW RAW RAW

EC NONE NONE NONE NONE NONE NONE PERCT NONE NONE PERCT DELTA NODISP NONE NONE NONE PERCT NONE PERCT NONE PERCT

FlexCache volume to origin Inode type stats on FlexCache volume Delegations on FlexCache volume Delegation fetch results on FlexCache volume CM_DIAG CM_DIAG CM_DIAG

Delegation return reasons CM_DIAG on FlexCache volume Delegation resets on FlexCache volume Delegation lookup queue length for hits Delegation lookup hit rate Delegation lookup operations Delegation lookup queue length for misses Delegation lookup miss rate Delegation lookup operations Inodes tossed during fill Data blocks tossed during fill Capacity of sparse volume Fullness of sparse volume Relative capacity of sparse volume Relative fullness of sparse volume Ideal # blocks used by volume in aggr rebalancing Pct of ideal # blocks used CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

delegation_looku

delegation_looku

delegation_looku

delegation_looku

Performance Management Design Guide

67 of 138

3/7/2008

vail_pct trunc_blks_unreclai mable trunc_blks_nblks

by volume in aggr rebalancing 8B RAW NONE Number of unreclaimable data blocks in FlexCache CM_DIAG volume Number of data blocks reclaimed by the truncator on FlexCache volume Number of inodes reclaimed by the truncator on FlexCache volume Max number of inodes in sparse volume's inofile Fullness of sparse volume's inofile CM_DIAG

8B

DELTA

NONE

trunc_blks_ninodes trunc_inos_max trunc_inos_usage trunc_inos_unreclai mable

8B 8B 8B 8B

DELTA RAW RAW RAW

NONE NONE PERCT NONE

CM_DIAG CM_DIAG CM_DIAG

Number of unreclaimable CM_DIAG inofile blocks on FlexCache volume Number of inofile blocks reclaimed by the truncator for sparseification on FlexCache volume Number of inodes reclaimed by the truncator for sparseification on FlexCache volume CM_DIAG

trunc_inos_nblks

8B

DELTA

NONE

trunc_inos_ninodes

8B

DELTA

NONE

CM_DIAG

fpolicy_global object Counter Name cifs_requests Size Type Display Unit NONE Description Filer-wide number of CIFS requests in FPolicy processing Filer-wide number of NFS requests in FPolicy processing Priority CM_DIAG Depends on

4B RAW

nfs_requests

4B RAW

NONE

CM_DIAG

Performance Management Design Guide

68 of 138

3/7/2008

fpolicy_stats_policy object Counter Name tlist_len tlist_max tlist_count tlist_recount Size Type Display Unit Description Length of the list of throttled requests Max length of the list of throttled requests Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

8B RAW NONE 8B RAW NONE

8B RAW NONE Total count of throttled requests Total count of the requests 8B RAW NONE returned to throttle list due to Servers being busy

fpolicy_stats_server object Counter Name avg_latency Size 8B Type Display Unit Description Priority CM_DIAG Depends on latency_base

Average latency for AVERAGE MSECS file policy operations in milliseconds Total time spent waiting for file policy requests to be used as a MSECS base counter for file policy average latency calculation Total no of requests send to server and NONE server is still processing them SECS Number of seconds since last screen completion Number of seconds since server was throttled (0 means not throttled)

latency_base

8B

DELTA

CM_DIAG

actv_reqs

8B

RAW

CM_DIAG

secs_since_last_ completion

8B

RAW

CM_DIAG

secs_since_throttle

8B

RAW

SECS

CM_DIAG

Performance Management Design Guide

69 of 138

3/7/2008

hostadapter object Counter Name total_reads Size 8B Type RATE Display Unit Description Priority CM_ADVANCED Depends on

Total number of PER_SEC reads on Host Adapter Total number of PER_SEC writes on Host Adapter PER_SEC PER_SEC Bytes reads via Host Adapter Bytes written via Host Adapter

total_writes bytes_read bytes_written

8B 8B 8B

RATE RATE RATE

CM_ADVANCED CM_ADVANCED CM_ADVANCED

ifnet object Counter Name recv_packets recv_errors send_packets send_errors collisions recv_data send_data recv_mcasts send_mcasts recv_drop_packet s Size 4B 4B 4B 4B 4B 8B 8B 4B 4B 4B Type Display Unit Description Packets received per second Errors per second while receiving packets Packets sent per second Errors per second while sending packets Collisions per second on CSMA interfaces Priority CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC Depends on

RAT PER_SE E C RAT PER_SE E C RAT PER_SE E C RAT PER_SE E C RAT PER_SE E C

RAT B_PER_S Total bytes received per E EC second RAT B_PER_S Total bytes sent per E EC second RAT PER_SE E C RAT PER_SE E C RAT PER_SE E C Packets received per second via multicast Packets sent per second via multicast Receive packets dropped per second

Performance Management Design Guide

70 of 138

3/7/2008

incpr_global object Counter Name totalpkts Size 8B Type DELTA Display Unit NONE Description Total Number of SpinNP packets for incremental copy restorer Priority CM_BASIC Depends on

committed inbound outbound

8B 8B 8B

RATE RATE RATE

Total KB committed to KB_PER_SEC disk by incremental copy restorer KB_PER_SEC KB_PER_SEC Total KB inbound to incremental copy restorer Total KB outbound from incremental copy restorer

CM_BASIC CM_BASIC CM_BASIC

waflwrites

8B

DELTA

NONE

Histogram of time in msec taken by wafl to write CM_BASIC blocks sent by incremental copy restorer Histogram of turn-around time in msec from incremental copy restorer Histogram of time spent by data packets in the backlog queue of incremental copy restorer Number of active incremental copy restore sessions Number of failed incremental copy restore sessions Number of incremental copy restore session creation failures due to too many sessions in progress CM_BASIC

turnaround

8B

DELTA

NONE

backedup

8B

DELTA

NONE

CM_TEST

active

8B

RAW

NONE

CM_BASIC

failed

8B

DELTA

NONE

CM_BASIC

hitmaxlimit

8B

DELTA

NONE

CM_BASIC

Performance Management Design Guide

71 of 138

3/7/2008

incpr_sess object Counter Name state vol_op Size 4B 4B Type RAW RAW Display Unit NONE NONE Description State of the Incremental Copy Restore Session Volume operation for the Incremental Copy Restore Session Transfer flags for the Incremental Copy Restore Session Time Elapsed since the Incremental Copy Restore started Total packets received by the Incremental Copy Restore Session Priority CM_DIAG CM_DIAG Depends on

xfer_flags

4B

RAW

NONE

CM_DIAG

elapsed

4B

RAW

MSECS

CM_DIAG

totalpkts

8B

RATE

PER_SEC

CM_DIAG

kb_committed

8B

RATE

Total kilobytes committed by the KB_PER_SEC Incremental Copy Restore Session Total kilobytes sent to KB_PER_SEC wafl by the Incremental Copy Restore Session Current backlog in the Incremental Copy Restore queue Current wafl message pool indicator in Incremental Copy Restore Session Total lock packets in Incremental Copy Restore Session Time spent in pre transfer phase for Incremental Copy Restore Session Time spent in data transfer phase for Incremental Copy Restore Session

CM_DIAG

kb_sent

8B

RATE

CM_DIAG

backlog

4B

DELTA NONE

CM_DIAG

pool_free

4B

RAW

NONE

CM_DIAG

lockpkts

4B

RATE

PER_SEC

CM_DIAG

pre_xfer_phase

4B

RAW

MSECS

CM_DIAG

data_xfer_phase 4B

RAW

MSECS

CM_DIAG

Performance Management Design Guide

72 of 138

3/7/2008

post_xfer_phase 4B

RAW

MSECS

Time spent in post transfer phase for Incremental Copy Restore Session Histogram of time in msec taken by wafl to write blocks sent by Incremental Copy Restorer

CM_DIAG

writes

8B

DELTA NONE

CM_BASIC

iomem object Counter Name cycles utilization read_ops Size 8B Type DELTA NODISP Display Unit DELTA NODISP PERCT PER_SEC Description Cycle counter for time base Percentage utilization of the I/O channel Number of read operations completed per second Average descriptor chain length per read operation Throughput for read operations per second Average read latency per block in microseconds Number of local read operations completed per second Average descriptor chain length per local read operation Priority CM_DIAG CM_DIAG CM_DIAG cycles Depends on

8B PERCT 8B RATE

read_chain read_thruput read_latency

8B AVERAGE NONE 8B RATE MB_PER_SEC

CM_DIAG CM_DIAG CM_DIAG

read_ops

8B AVERAGE USECS

read_ops

lread_ops

8B RATE

PER_SEC

CM_DIAG

lread_chain

8B AVERAGE NONE

CM_DIAG

lread_ops

lread_thruput lread_latency

8B RATE

Throughput for local MB_PER_SEC read operations per second. Average local read latency per block in

CM_DIAG CM_DIAG lread_ops

8B AVERAGE USECS

Performance Management Design Guide

73 of 138

3/7/2008

microseconds write_ops 8B RATE PER_SEC Number of write operations completed per second Average descriptor chain length per write operation Throughput for write operations per second. Average write latency in microseconds Number of local write operations completed per second Average descriptor chain length per local write operation CM_DIAG

write_chain write_thruput write_latency lwrite_ops

8B AVERAGE NONE 8B RATE MB_PER_SEC

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

write_ops

8B AVERAGE USECS 8B RATE PER_SEC

write_ops

lwrite_chain

8B AVERAGE NONE

CM_DIAG

lwrite_ops

lwrite_thruput 8B RATE

Throughput for local MB_PER_SEC write operations per second. Average local write latency per block in microseconds Total number of unrecoverable DMA errors Total number of descriptor CRC errors Total number of data CRC errors Total number of correctable ECC memory errors Total number of uncorrectable ECC memory errors Total number of I/O operations received Total number of I/O operations completed

CM_DIAG

lwrite_latency 8B AVERAGE USECS

CM_DIAG

lwrite_ops

dma_errs desc_crc_errs data_crc_errs cecc_errs

8B RAW 8B RAW 8B RAW 8B RAW

NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

uecc_errs received completed

8B RAW 8B RAW 8B RAW

NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

74 of 138

3/7/2008

ldap object Counter Name Size avg_latency 8B Type Display Unit Description Average latency for ldap operations in milliseconds Priority CM_DIAG Depends on latency_base

AVERAGE MSECS

latency_base

8B

DELTA

Total time spent waiting for ldap requests to be used as a NONE CM_DIAG base counter for ldap average latency calculation

lmem object Counter Name data_mapped io_data_mapped data2_mapped data_stripped io_data_stripped data2_stripped indirect_stripped suspended suspended_in_scv nosusp_pre_valloc restarted scavenge_okay scavenge_failed cp_from_low_vbuf nosusp_no_surplus dirty_cnt_sflush dirty_single_flush Size 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 4B 8B 8B 8B Type RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW Display Unit NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE Description data_mapped io_data_mapped data2_mapped data_stripped io_data_stripped data2_stripped indirect_stripped suspended suspended_in_scv nosusp_pre_valloc restarted scavenge_okay scavenge_failed cp_from_low_vbuf no_surplus_from_cp nosusp_no_surplus dirty_cnt_sflush dirty_single_flush Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

no_surplus_from_cp 8B

Performance Management Design Guide

75 of 138

3/7/2008

dirty_batch_flush dirty_global_flush child_temp_mapped no_child_buf no_child_map

8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE

dirty_batch_flush dirty_global_flush child_ret_unmapped child_temp_mapped no_child_buf no_child_map

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

child_ret_unmapped 8B

lmgr_ng object Counter Name locks locks_max locks_limit locks_byte locks_share locks_granted Size 4B 4B 4B 4B 4B 4B Type Display Unit Description Number of allocated lock objects in the system Priority CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED Depends on

RAW NONE

Highest number of lock RAW NONE objects in the system at one time so far RAW NONE RAW NONE RAW NONE Upper limit on the number of lock objects in the system Number of byte lock objects in the system Number of share lock objects in the system

Number of lock objects in RAW NONE the system which are granted Number of lock objects in RAW NONE the system waiting to be granted RAW NONE Number of allocated file objects in the system

locks_waiting files files_max files_limit hosts hosts_max

4B 4B 4B 4B 4B 4B

CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED

Highest number of file RAW NONE objects in the system at one time so far RAW NONE RAW NONE Upper limit on the number of file objects in the system Number of allocated host objects in the system

RAW NONE Highest number of host

Performance Management Design Guide

76 of 138

3/7/2008

objects in the system at one time so far hosts_limit owners owners_max 4B 4B 4B RAW NONE RAW NONE Upper limit on the number of host objects in the system Number of allocated owner objects in the system CM_ADVANCED CM_ADVANCED CM_ADVANCED

Highest number of owner RAW NONE objects in the system at one time so far Upper limit on the number RAW NONE of owner objects in the system

owners_limit

4B

CM_ADVANCED

logical_replication_destination object Counter Name num_transfers num_success_transfers Size 4B 4B Type RAW RAW RAW RAW RAW RAW RAW STRIN G RAW STRIN G STRIN Display Unit NONE NONE NONE NONE NONE NONE NONE NONE SECS NONE NONE Description Number of transfers since reboot Number of successful transfers since reboot Number of consecutive successful transfers since last failure Number of transfer failures since reboot Number of consecutive transfer failures since last success Number of times the transfer was restarted Total number of different types of files processed since reboot Current state of the replica Current lag time of the replica Name of the base snapshot for the last transfer Last transfer type Priority CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC

Dep

num_consecutive_succes 4B ses num_failures num_consecutive_failure s num_restarts total_num_files state lag_time last_base_snapshot last_transfer_type 4B 4B 4B 4B 32B 4B 160B 32B

Performance Management Design Guide

77 of 138

3/7/2008

G last_transfer_size last_transfer_duration transfer_status last_transfer_cpu_usage check_point_num transfer_phase num_headers_read num_data_blocks_read network_data_read network_read_rate num_inode_deletes directory_phase_time num_directory_phase_bl ocks 8B 4B 16B 8B 4B 32B 4B 8B 8B 8B 4B 8B 4B RAW RAW STRIN G RAW RAW STRIN G RAW RAW RAW RATE RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE SECS NONE PERCT NONE NONE NONE NONE NONE Amount of data transferred during last transfer in bytes Time taken for the last transfer Status of the current transfer Percentage of CPU used during last transfer Last checkpoint number Current phase of the transfer (valid only for active transfer) Number of header blocks read from network Number of data blocks read from network Number of bytes read from network CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

KB_PER_ Network read rate during SEC transfer NONE MSECS NONE NONE NONE NONE NONE NONE NONE NONE Number of inodes deleted since last transfer Total time spent in directory phase receiving data from source Number of blocks received in directory phase Number of chunks received in directory phase Number of stream directory chunks received Number of acl delete chunks received Number of stream directory delete chunks received Number of inode delete chunks received Number of directory chunks received Number of special files chunks

num_directory_phase_ch 4B unks num_streamdir_chunks num_acldel_chunks num_streamdirdel_chun ks num_true_delete_chunks num_dir_chunks num_spcl_file_chunks 4B 4B 4B 4B 4B 4B

Performance Management Design Guide

78 of 138

3/7/2008

received num_acl_data_chunks num_extattr_chunks num_discard_data_chun ks num_mid_stream_chunk s dir_processing_time num_direntry_deletes 4B 4B 4B 4B 8B 4B RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE MSECS NONE NONE NONE NONE NONE MSECS NONE NONE MSECS NONE Number of acl chunks received Number of extended attributes chunks received Number of discard data chunks received Number of mid stream chunks received CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Total time spent in the directory CM_DIAG processing phase Number of directory entries deleted Number of directory entries deleted due to renames Number of directory entries created Number of directory entries created due to renames Number of blocks read from temporary files during directory processing Time taken for the file phase Total blocks received in file phase Number of times the network read thread waited on full prefetch queue Total time network read thread waited on full prefetch queue Number of times the prefetch thread waited for the network read thread to provide data Total time prefetch thread waited for the network thread to provide data Number of times datawrite thread waited for the prefetch thread to provide data Total time datawrite thread CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

num_direntry_delete_ren 4B ames num_direntry_creates 4B

num_direntry_create_ren 4B ames dir_processing_read_req file_phase_time file_phase_blocks num_prefetch_q_full 4B 8B 8B 4B

wait_time_prefetch_q_fu 8B ll num_prefetch_q_empty wait_time_prefetch_q_e mpty num_datawrite_q_empty 4B

8B

RAW

MSECS

CM_DIAG

4B

RAW RAW

NONE MSECS

CM_DIAG CM_DIAG

wait_time_datawrite_q_e 8B

Performance Management Design Guide

79 of 138

3/7/2008

mpty num_inofile_blocks_pref 4B etched num_file_blocks_prefetc hed num_file_blocks_written num_inomap_entry_gen _adds 8B 8B 8B RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE MSECS MSECS MSECS MSECS MSECS MSECS MSECS MSECS

waited for the prefetch thread to provide data Number of inode file blocks prefetched Number of file blocks prefetched Number of file data blocks written Number of generation entries added to inodemap Number of inode mapping entries added to inodemap Number of inodemap entries deleted Number of inodemap entries read from inodemap Number of streamdir entries set in the inodemap Number of streamdir entries cleared from inodemap Number of acl flags set in the inodemap Number of acl flags cleared from inodemap Total CPU time used by all threads for the last transfer Total time used by all threads for the last transfer CPU time used by cleantree threads Total time used by cleantree threads CPU time used by buildtree threads Total time used by buildtree threads CPU time used by prefetch threads Total time used by prefetch CM_DIAG CM_DIAG CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

num_inomap_entry_fh_a 8B dds num_inomap_entry_dele tes num_inomap_entry_read s num_inomap_strmdir_se t num_inomap_strmdir_cl r num_inomap_acl_set num_inomap_acl_clr total_cpu_time total_time 8B 8B 8B 8B 8B 8B 8B 8B

cleantree_process_cpu_ti 8B me cleantree_process_time buildtree_process_cpu_ti me buildtree_process_time prefetch_process_cpu_ti me prefetch_process_time 8B 8B 8B 8B 8B

Performance Management Design Guide

80 of 138

3/7/2008

threads datawrite_process_cpu_t ime datawrite_process_time main_proc_process_cpu _time main_proc_process_time 8B 8B 8B 8B RAW RAW RAW RAW MSECS MSECS MSECS MSECS CPU time used by datawrite threads Total time used by datawrite threads CPU time used by the main process Total time used by main process Total number of new or modified files processed during current transfer CM_DIAG CM_DIAG CM_DIAG CM_DIAG

num_files

4B

RAW

NONE

CM_BASIC

Performance Management Design Guide

81 of 138

3/7/2008

logical_replication_source object Counter Name num_transfers num_cksum_blk_packed num_success_transfers Size 4B 4B 4B Type RAW RAW RAW RAW RAW RAW RAW RAW RAW RATE RAW RAW RAW RAW STRI NG RAW RAW RAW STRI NG RAW STRI NG RAW RAW RAW RAW RAW STRI NG RAW RAW RAW RAW STRI RAW NG RAW RAW RAW RAW Display Unit NONE NONE NONE NONE NONE Description Priority Depends on

Number of transfers since CM_BASIC reboot Number of checksum blocks Number of successful saved by multiple file packing CM_DIAG CM_BASIC transfers since reboot in headers CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_BASIC CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

num_data_tinyfile_packe num_consecutive_success 4B 4B d es num_data_blocks_written num_failures network_data_write 8B 4B 8B

num_consecutive_failures 4B network_write_rate 8B num_restarts num_inode_deletes dir_phase_time total_num_files file_phase_time state inofile_blocks_prefetched lag_time inofile_blocks_prefetch_ msgs last_base_snapshot inofile_blocks_processed last_transfer_type inodes_processed last_transfer_size file_blocks_prefetched last_transfer_duration dir_blocks_prefetched transfer_status acl_blocks_prefetched last_transfer_cpu_usage num_pass1_empty_wait check_point_num transfer_phase time_pass1_empty_wait num_headers_written num_pass1_full_wait num_headers_packed time_pass1_full_wait 4B 4B 8B 4B 8B 32B 4B 4B 4B 160B 4B 32B 8B 8B 8B 4B 8B 16B 4B 8B 4B 4B 32B 8B 4B 4B 4B 8B

consecutive Number of blocks saved by successful transfersheaders tinyfiles packed in since last failure Number of data blocks NONE Number the network written to of transfer failures NONE since reboot Number of bytes written to NONE Number network of consecutive NONE transfer failures since last KB_PER_ Network write rate during success SEC transfer Number of times the transfer Number of inodes deleted NONE NONE was restarted since last transfer Total number of different Total time taken in directory MSECS types of files transferred since NONE phase reboot Total time taken in file MSECS phase NONE Current state of the replica NONE SECS NONE NONE NONE NONE NONE NONE NONE SECS NONE NONE NONE PERCT NONE NONE NONE MSECS NONE NONE NONE MSECS Inode file blocks prefetched Current lag time of the Number replica of prefetch messages to read inode file Name of the base snapshot blocks for the last transfer Number of inode file blocks processed Last transfer type Number of inodes processed Amount of data transferred Number of blocks during last transfer in bytes prefetched in file phase Time taken for the last Number transfer of blocks prefetched in directory phase Status of the current transfer Number of acl blocks prefetched Percentage of CPU usage for Number of times the last transfer worker threads waited on empty Last checkpoint number pass1 queue Current phase of the transfer Total time worker threads (valid onlyempty pass1 queue waited on for active transfer) Number of header blocks Number of times inode written to the network picker thread waited on full Number of pass1 queue header blocks saved by multiple file packing Total time inode picker in headers

Performance Management Design Guide

82 of 138

3/7/2008

thread waited on full pass1 queue num_pass2_empty_wait time_pass2_empty_wait num_pass2_full_wait 4B 8B 4B RAW RAW RAW NONE MSECS NONE Number of times worker threads waited on empty pass2 queue Total time worker threads waited on empty pass2 queue Number of times inode picker thread waited on full pass2 queue Total time inode picker thread waited on full pass2 queue Number of inode file block reads issued by inode picker Number of new files created since last transfer Number of files modified since last transfer Number of regular files deleted since last transfer Number of directories deleted since last transfer CM_DIAG CM_DIAG CM_DIAG

time_pass2_full_wait num_inofile_blocks_read num_new_files num_update_files num_deleted_files num_deleted_dirs num_acl_deletes num_ntstreamdir_rels num_streamdir_deletes num_maps_processed num_holes num_file_blocks_read num_dir_blocks_read total_cpu_time total_time pass1_inode_picker_proc

8B 8B 4B 4B 4B 4B 4B 4B 4B 4B 8B 8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

MSECS NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE MSECS MSECS MSECS

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Number of acls deleted since CM_DIAG last transfer Number of NT stream relationships transferred Number of stream directory deletes since last transfer Number of maps processed Number of holes transferred Number of file blocks read Number of directory blocks read Total CPU time used by all threads for the last transfer Total time used by all threads for the last transfer CPU time used by pass1 CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC CM_BASIC CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

83 of 138

3/7/2008

ess_cpu_time pass1_inode_picker_proc ess_time pass2_inode_picker_proc ess_cpu_time pass2_inode_picker_proc ess_time worker_process_cpu_tim e worker_process_time pipeline_process_cpu_tim e pipeline_process_time BLI_process_cpu_time BLI_process_time 8B 8B 8B 8B 8B 8B 8B 8B 8B RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW MSECS MSECS MSECS MSECS MSECS MSECS MSECS MSECS MSECS MSECS MSECS NONE

inode picker thread Total time used by pass1 inode picker thread CPU time used by pass2 inode picker thread Total time used by pass2 inode picker thread CPU time used by worker threads Total time used by worker threads CPU time used by pipeline thread Total time used by pipeline thread CPU time used by block level incremental thread Total time used by block level incremental thread CPU time used by the main process Total time used by the main process Total number of new or modified files transferred during current transfer CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC

main_proc_process_cpu_t 8B ime main_proc_process_time num_files 8B 4B

ndmp object Counter Name dir_buffers_sent Size 4B Type RAW Display Unit NONE Description Number of buffers of DIR file history sent to NDMP Number of buffers of NODE file history sent to NDMP Number of times DIR Priority CM_BASIC Depends on

node_buffers_sent dir_send_was_blocked

4B 4B

RAW RAW

NONE NONE

CM_BASIC CM_ADVA

Performance Management Design Guide

84 of 138

3/7/2008

file history send to NDMP was blocked node_send_was_blocke d dir_flush_calls node_flush_calls num_node_entries num_dir_entries num_dir_entries_2fh dir_entry_2fh_min_late ncy dir_entry_2fh_max_late ncy 4B 4B 4B 8B 8B 8B 8B 8B RAW RAW RAW RAW RAW RAW RAW RAW RAW NONE NONE NONE NONE NONE NONE MSECS MSECS MSECS

NCED

Number of times CM_ADVA NODE file history send NCED to NDMP was blocked Number of DIR file history flush operations Number of NODE file history flush operations Number of NODE file history entries Number of DIR file history entries Number of DIR file history entries to FH CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC

Min xmit time for DIR CM_ADVA file history entries to FH NCED Max xmit time for DIR CM_ADVA file history entries to FH NCED Average xmit time for CM_ADVA DIR file history entries NCED to FH Total xmit time for DIR file history entries to FH CM_ADVA NCED

dir_entry_2fh_ave_laten 8B cy dir_entry_2fh_tot_laten cy num_node_entries_2fh node_entry_2fh_min_la tency node_entry_2fh_max_la tency node_entry_2fh_ave_lat ency 8B 8B 8B

RAW RAW RAW

MSECS NONE MSECS

Number of NODE file CM_ADVA history entries to FH NCED Min xmit time for NODE file history entries to FH Max xmit time for NODE file history entries to FH CM_ADVA NCED CM_ADVA NCED

8B

RAW

MSECS

8B

RAW

MSECS

Average xmit time for CM_ADVA NODE file history NCED entries to FH Total xmit time for NODE file history entries to FH Number of DIR file CM_ADVA NCED CM_BASIC

node_entry_2fh_tot_late 8B ncy num_dir_entries_2ndmp 8B

RAW RAW

MSECS NONE

Performance Management Design Guide

85 of 138

3/7/2008

history entries to NDMP dir_entry_2ndmp_min_l atency dir_entry_2ndmp_max_ latency dir_entry_2ndmp_ave_l atency dir_entry_2ndmp_tot_la tency num_node_entries_2nd mp 8B RAW MSECS Min xmit time for DIR file history entries to NDMP CM_BASIC

8B

RAW

MSECS

Max xmit time for DIR file history entries to CM_BASIC NDMP Average xmit time for DIR file history entries to NDMP Total xmit time for DIR file history entries to NDMP Number of NODE file history entries to NDMP Min xmit time for NODE file history entries to NDMP Max xmit time for NODE file history entries to NDMP Average xmit time for NODE file history entries to NDMP Total xmit time for NODE file history entries to NDMP Max depth for File History queue Number of times File History queue was full CM_BASIC

8B

RAW

MSECS

8B

RAW

MSECS

CM_BASIC

8B

RAW

NONE

CM_BASIC

node_entry_2ndmp_min 8B _latency node_entry_2ndmp_ma x_latency node_entry_2ndmp_ave _latency 8B

RAW

MSECS

CM_BASIC

RAW

MSECS

CM_BASIC

8B

RAW

MSECS

CM_BASIC

node_entry_2ndmp_tot_ 8B latency max_queue_depth fh_queue_full_cnt 4B 4B

RAW RAW RAW

MSECS NONE NONE

CM_BASIC CM_ADVA NCED CM_ADVA NCED

nfsv3 object Counter Name nfsv3_ops Size 8B Type RATE Display Unit PER_SEC Description Total number of NFS v3 operations per second 86 of 138 Priority CM_BASIC

Depend

Performance Management Design Guide

3/7/2008

nfsv3_read_laten cy nfsv3_avg_read_ latency_base nfsv3_read_ops nfsv3_read_laten cy_hist nfsv3_write_late ncy nfsv3_avg_write _latency_base nfsv3_write_ops nfsv3_write_late ncy_hist nfsv3_op_count nfsv3_op_latenc y_base nfsv3_op_percen t nfsv3_op_latenc y

8B 8B 8B 8B

AVERAGE DELTA NODISP RATE DELTA

MSECS DELTA NODISP PER_SEC NONE

Average latency for NFS v3 read operations in milliseconds Number of NFS V3 reads for latency calculation Total observed NFS V3 read operations per second Histogram of latency for NFS V3 read operations in milliseconds Average latency for NFS v3 write operations in milliseconds Number of NFS V3 writes for latency calculation Total observed NFS v3 write operations per second

CM_BASIC CM_BASIC CM_BASIC CM_DIAG

nfsv3_avg_read_lat

8B

AVERAGE DELTA NODISP RATE DELTA DELTA DELTA NODISP PERCT AVERAGE DELTA DELTA AVERAGE DELTA NODISP

MSECS DELTA NODISP PER_SEC NONE NONE DELTA NODISP PERCT USECS NONE NONE USECS DELTA NODISP

CM_BASIC

nfsv3_avg_write_la

8B 8B 8B 8B 8B

CM_BASIC CM_BASIC

Histogram of latency for NFS V3 write operations in CM_DIAG milliseconds Array of select NFS v3 operation counts Array of select NFS v3 operation counts for latency calculation Array of select NFS v3 operations as a percentage of total NFS v3 operations Array of latencies of select NFS v3 operations Histogram of NFS v3 read sizes Histogram of NFS v3 write sizes CM_DIAG CM_DIAG

8B 8B

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

nfsv3_ops

nfsv3_op_latency_b

nfsv3_read_size_ 4B histo nfsv3_write_size _histo nfsv3_avg_op_la tency nfsv3_avg_op_la tency_base 4B 8B 8B

Average latency of the nfs CM_DIAG V3 ops Number of NFS V3 ops for average latency calculation CM_DIAG

nfsv3_avg_op_laten

Performance Management Design Guide

87 of 138

3/7/2008

nfsv3_latency_hi st

8B

DELTA

NONE

Histogram of latency for NFS V3 operations in milliseconds

CM_DIAG

nfsv4 object Counter Name nfsv4_calls Size 8B Type RATE Display Unit Description Priority CM_DIAG Depends on

Number of NFS v4 calls PER_SEC (null or compound) per second NONE PER_SEC NONE Number of NFSV4 NULL calls and COMPOUND calls Number of NFS v4 compound calls per second Array of select NFS v4 operation counts Array of percent of select NFS v4 operations calculated as \ percent of total nfsv4 operations Number of total NFS v4 operations per second Average latency for NFS v4 operations in milliseconds Histogram of latency for NFS V4 operations in milliseconds Number of NFS v4 read operations per second Average latency for NFS v4 Read operations in milliseconds Histogram of latency for NFS V4 operations in milliseconds Number of NFS v4 write operations per second 88 of 138

nfsv4_calls_types nfsv4_cmpnds nfsv4_op_count

8B 8B 8B

DELTA RATE DELTA

CM_DIAG CM_DIAG CM_DIAG

nfsv4_op_percent

8B

PERCT

PERCT

CM_DIAG

nfsv4_ops

nfsv4_ops nfsv4_ops_latency

8B 8B

RATE AVERA GE DELTA RATE AVERA GE DELTA RATE

PER_SEC MSECS

CM_DIAG CM_BASIC nfsv4_ops

nfsv4_latency_hist nfsv4_read_ops nfsv4_read_latency nfsv4_read_latency_ hist nfsv4_write_ops

8B 8B 8B

NONE PER_SEC MSECS

CM_DIAG CM_DIAG CM_BASIC nfsv4_read_ops

8B 8B

NONE PER_SEC

CM_DIAG CM_DIAG

Performance Management Design Guide

3/7/2008

nfsv4_write_latency

8B

AVERA GE DELTA

MSECS

Average latency for NFS v4 Write operations in milliseconds Histogram of latency for NFS V4 Write operations in milliseconds Number of times no delegation was handed out to clients \ because of some error Number of times read delegation was handed out to clients Number of times write delegation was handed out to clients Array of NFS v4 reply cache opinfo Array of V4 reply cache reply Average latency of a compound request

CM_BASIC nfsv4_write_ops

nfsv4_write_latency_ 8B hist

NONE

CM_DIAG

nfsv4_nodelegation

4B

DELTA

NONE

CM_DIAG

nfsv4_read_delegatio n nfsv4_write_delegati on nfsv4_reply_cache_o pinfo nfsv4_reply_cache_r eply nfsv4_cmpnd_latenc y nfsv4_cmpnd_latenc y_base nfsv4_op_latency nfsv4_op_latency_ba se nfsv4_read_size_hist o nfsv4_write_size_his to nfsv4_avg_latency nfsv4_avg_latency_b ase

4B

DELTA

NONE

CM_DIAG

4B 4B 4B 8B 8B 8B 8B 8B 8B 8B 8B

DELTA RAW RAW AVERA GE DELTA NODIS P AVERA GE DELTA NODIS P DELTA DELTA AVERA GE DELTA NODIS P

NONE NONE NONE USECS DELTA NODISP USECS DELTA NODISP NONE NONE USECS DELTA NODISP

CM_DIAG CM_DIAG CM_DIAG CM_DIAG nfsv4_cmpnd_lat ency_base

Number of compounds for CM_DIAG average latency computation Array of latencies of select NFS v4 operations Array of select NFS v4 ops counts for latency calculation Histogram of NFS v4 read sizes Histogram of NFS v4 write sizes Average latency of an NFSv4 request Number of total NFSv4 ops for average latency computation CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG nfsv4_avg_latenc y_base nfsv4_op_latenc y_base

Performance Management Design Guide

89 of 138

3/7/2008

nrv object Counter Name reqs_enobufs Size 8B Type DELTA Display Unit NONE Description Number of failed NRV requests with ENOMEM in nrv server Number of failed NRV requests which needed initialization in nrv server Total number of NRV requests deferred in nrv server Total number of NRV requests deferred in nrv server due to full send buffer Total number of times we entered read defer mode in NRV Total number of NRV responses sent in nrv server Number of NRV responses currently deferred in nrv server Total number of NRV responses deferred in nrv server Number of NRV hearbeats sent in nrv server Number of NRV heartbeats received in nrv server Number of NRV Priority CM_DIAG Depends on

reqs_needs_init

8B

DELTA

NONE

CM_DIAG

reqs_deferred

8B

DELTA

NONE

CM_DIAG

reqs_deferred_sendb uf

8B

DELTA

NONE

CM_DIAG

reqs_entered_read_d efer

8B

DELTA

NONE

CM_DIAG

resp_sent

8B

DELTA

NONE

CM_DIAG

resp_curr_deferred

8B

RAW

NONE

CM_DIAG

resp_deferred

8B

DELTA

NONE

CM_DIAG

htbeat_sent

8B

DELTA

NONE

CM_DIAG

htbeat_received htbeat_connclosed

8B 8B

DELTA DELTA

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

90 of 138

3/7/2008

connections closed by heartbeat monitor in nrv server conn_active 8B RAW NONE Number of active NRV connections in nrv server Number of closed NRV connections in nrv server Number of read transfers requests received in nrv server Number of blocks that received read requests in nrv server Number of readahead transfer requests received in nrv server Number of blocks that received readhead requests in nrv server CM_DIAG

conn_closed

8B

DELTA

NONE

CM_DIAG

read_xfers

8B

DELTA

NONE

CM_DIAG

read_blocks

8B

DELTA

NONE

CM_DIAG

readahead_xfers

8B

DELTA

NONE

CM_DIAG

readahead_blocks

8B

DELTA

NONE

CM_DIAG

nvram object Counter Name total_nvlog_data write_data Size 8B 8B Type RATE RATE Display Unit Description Priority CM_DIAG CM_DIAG Depends on B_PER_SE Total nvlog bytes C per second NVRAM data in B_PER_SE bytes written per C second NVRAM data in B_PER_SE bytes DMAed per C second MSECS NVRAM DMA wait time per transaction

total_dma_data

8B

RATE

CM_DIAG

dma_wait_latency

8B

AVERAGE

CM_DIAG

transaction_count

Performance Management Design Guide

91 of 138

3/7/2008

transaction_count nvlog_header_array_ full

8B 8B

RATE RAW

PER_SEC NONE

NVRAM DMA transactions per second NVLOG shadow header full count

CM_DIAG CM_DIAG

partial_file_restore object Counter Name requests bytes_restored Size 8B 8B Type Display Unit Description Partial file restore requests per second Partial file restore bytes restored per second Priority CM_DIAG CM_DIAG Depends on

RATE PER_SEC RATE PER_SEC

priorityqueue object Counter Name weight usr_weight usr_sched_total usr_pending avg_usr_pending_ ms usr_queued_total sys_sched_total sys_pending avg_sys_pending_ ms Size 4B 4B 8B 8B 8B 8B 8B 8B 8B Type RAW RAW RATE RAW AVERA GE RATE RATE RAW AVERA GE Display Unit NONE NONE PER_SEC NONE MSECS PER_SEC PER_SEC NONE MSECS Description Queue scheduling weight User request scheduling weight within a queue Number of scheduling user requests Number of pending user requests Priority CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED Depends on

Average pending time for CM_ADVA usr_queued_tota usr messages in NCED l milliseconds Number of user requests placed on queue Number of scheduled system requests Number of pending system requests CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED

Average pending time for CM_ADVA sys_queued_tota system messages in NCED l 92 of 138 3/7/2008

Performance Management Design Guide

milliseconds sys_queued_total usr_read_limit 8B 4B RATE RAW PER_SEC NONE Number of requests placed on queue Limit on user reads outstanding on this queue CM_ADVA NCED CM_ADVA NCED

max_user_reads

4B

RAW

NONE

Maximum number of user reads ever CM_ADVA outstanding at one time for NCED this queue Limit on system reads outstanding on this queue CM_ADVA NCED

sys_read_limit

4B

RAW

NONE

max_sys_reads

4B

RAW

NONE

Maximum number of system read requests ever CM_ADVA outstanding at one time for NCED this queue Number of times the user CM_ADVA read limit was hit NCED Number of times the system read limit was hit Amount of nvlog this queue may use during CP Maximum amount of nvlog this queue used during a CP Number of times the nvlog queue limit was hit during CP Number of times the queue was not usable during schedule dequeue Number of times the queue was a candidate for dequeue Number of times a delayed system message was pushed Number of times a delayed user message was pushed Number of times this CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED CM_ADVA NCED CM_DIAG

usr_read_limit_hit sys_read_limit_hit nvlog_limit nvlog_used_max

8B 8B 4B 4B

DELTA DELTA RAW RAW

NONE NONE NONE NONE

nvlog_limit_full

4B

DELTA

NONE

queue_not_usable queue_maybe_cho ose sys_delayed

8B

RAW

NONE

8B

RAW

NONE

CM_DIAG

8B

RAW

NONE

CM_DIAG

usr_delayed first_hit

8B 8B

RAW RAW

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

93 of 138

3/7/2008

queue was hit first retry_hit cand_hit 8B 8B RAW RAW NONE NONE Number of times this queue was hit on retry Number of times this queue was used from candidate queues Number of times this queue was a candidate during retry Number of times this queue found a user message that was too oldwhen compared to system messages Number of times this queue found a system message that was too oldwhen compared to user messages CM_DIAG CM_DIAG

retry_cand_hit

8B

RAW

NONE

CM_DIAG

usr_old

8B

DELTA

NONE

CM_DIAG

sys_old

8B

DELTA

NONE

CM_DIAG

prisched object Counter Name queued queued_max wake_on_sig_rate Size 4B 4B 4B Type RAW RAW RATE RATE RATE RATE Display Unit NONE NONE PER_SEC PER_SEC PER_SEC Description Current messages queued Maximum schduling queue depth Scheduler wake on signal received rate Scheduler wake on message received rate Scheduler preemption rate Priority CM_ADVANCED CM_ADVANCED CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

wake_on_mesg_rate 4B preempt_rate bypass_rate 4B 4B

Rate at which PER_SEC guardrail scheduler is bypassed. NONE Number of times a dequeue operation retried due to range

retry_range

4B

RAW

CM_DIAG

Performance Management Design Guide

94 of 138

3/7/2008

readjustment nvlog_threshold 4B RAW NONE NVLOG used limit before enforcing perqueue limits Number of times scheduler was found to be slow. Number of times a delayed message was found and pushed. Number of times delayed messages were found and all queues were I/O blocked. Number of attemped asynchronous dispatches. Number of attemped asynchronous dispatches that would block CM_DIAG

schedslow

4B

RAW

NONE

CM_DIAG

delayed_messages

4B

RAW

NONE

CM_DIAG

delayed_io_blocked

4B

RAW

NONE

CM_DIAG

try_send

4B

RAW

NONE

CM_DIAG

try_send_block

4B

RAW

NONE

CM_DIAG

qtree object Counter Name parent_vol nfs_ops cifs_ops Size Type Display Unit Description Name of the parent volume Number of NFS operations per second to the qtree Number of CIFS operations per second to the qtree Priority CM_DIAG CM_BASIC CM_BASIC Depends on

64B STRING NONE 8B 8B RATE RATE PER_SEC PER_SEC

internal_ops 8B

RATE

Number of internal operations generated by PER_SEC activites such as snapmirror and backup per second to the qtree

CM_DIAG

Performance Management Design Guide

95 of 138

3/7/2008

quota object Counter Name quota_state Size 4B Type RAW Display Unit NONE Description Priority

Quota state: 0=off CM_PRIV_DIAG|CM_ASUP_EVENT|CM 1=on 2=init T

quota_fsi_state

4B

RAW

NONE

Quota state in volinfo: 0=off 1=on 2=init 3=shutdown CM_PRIV_DIAG|CM_ASUP_EVENT|CM 4=enable 5=disable T 6=maint_off 7=snap_off. See wafl_Quota_state Quota database file size in blocks

quota_db_blocks quota_name_db_bl ocks quota_records quota_disk_record s quota_types quota_lookups

4B 4B 4B 4B 4B 4B

RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE

CM_PRIV_DIAG|CM_ASUP_EVENT|CM T

Quota name CM_PRIV_DIAG|CM_ASUP_EVENT|CM database file size in T blocks Quota records in memory Quota disk records

CM_PRIV_DIAG|CM_ASUP_EVENT|CM T

CM_PRIV_DIAG|CM_ASUP_EVENT|CM T

Quota disk record CM_PRIV_DIAG|CM_ASUP_EVENT|CM T typess Various stats for Quota lookups

CM_PRIV_DIAG|CM_ASUP_EVENT|CM T

readahead object Counter Name blks_read blks_async blks_ext_cache_hit chains_read_ahead dummy_reads requested Size 8B 8B 8B 8B 8B 8B Type DELTA DELTA DELTA DELTA DELTA DELTA Display Unit NONE NONE NONE NONE NONE NONE Description Total blocks read Async blocks Blocks hit in external cache Chains read via readahead Dummy reads in readahead Blocks of readahead Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

Performance Management Design Guide

96 of 138

3/7/2008

requests read incore past_eof hole absent noabsent noblocks nomsgs nobufs noraidbufs speculative noreadsets read_once dir_blocks spanned readsets_assign readsets_release readsets_notused total_read_reqs 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE Blocks actually read Blocks already resident Blocks requested past EOF CM_DIAG CM_DIAG CM_DIAG

Blocks requested from CM_DIAG holes Blocks requested from CM_DIAG absents Request absent blocks what were skipped CM_DIAG

Blocks requested from CM_DIAG lev-0 inodes Blocks dropped for lack of messages Blocks dropped for lack of buffers Blocks dropped for lack of RAID buffers Blocks speculatively read No readsets available Read-once blocks Directory blocks read ahead Spanned read aheads Readsets assigned Readsets released Readsets not used Histogram of total read requests seen by the RA algorithm Histogram of sequential read request percentages as seen by the readahead algorithm CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

seq_read_reqs

8B

PERCT

PERCT

CM_DIAG

total_read_reqs

Performance Management Design Guide

97 of 138

3/7/2008

rand_read_reqs

8B

PERCT

PERCT

Histogram of random read request percentages as seen by the readahead algorithm Histogram of readahead chains sent to RAID Readahead rescans due to read sequence break Previous readset reused had a short count

CM_DIAG

total_read_reqs

chain_lengths

4B

DELTA

NONE

CM_DIAG

rb_rescan

8B

DELTA

NONE

CM_DIAG

prev_short

8B

DELTA

NONE

CM_DIAG

rs_max_counts prefetch_requests prefetch_hit prefetch_hit_valid prefetch_ibuf prefetch_noparent

4B 8B 8B 8B 8B 8B

DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA

Histogram of NON_ZERO maximum readset counts when recycled NONE NONE NONE NONE NONE NONE NONE Readahead prefetch requests prefetch buf load hits prefetch buf load hitsvalid buffer prefetch buffer in inode prefetch lacked parent buffer prefetch buffer is decompressing attempt to prefetch a hole attempt to prefetch a buffer when pvbn conversion not available attempt to prefetch when no buffers available prefetch attempted read prefetch hit in

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

prefetch_decompress 8B prefetch_hole 8B

prefetch_nopvn

8B

DELTA

NONE

CM_DIAG

prefetch_nobufs prefetch_read prefetch_ec_hit

8B 8B 8B

DELTA DELTA DELTA

NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

98 of 138

3/7/2008

extended cache prefetch_read_type ext_cache_hit 8B 8B DELTA DELTA NON_ZERO read types of reads NONE readahead hit in extended cache number of times cached read optimization was selected Blocks skipped due to fake-dirty parent write Read reallocation requested on all blocks of a readahead Read reallocation requested on some blocks of a readahead Read reallocation requested on no blocks of a readahead CM_DIAG CM_DIAG

cached_read_opt

8B

RATE

PER_SEC

CM_DIAG

no_fd_parent rr_all

8B 8B

RATE DELTA

PER_SEC NONE

CM_DIAG CM_DIAG

rr_small

8B

DELTA

NONE

CM_DIAG

rr_none

8B

DELTA

NONE

CM_DIAG

raid object Counter Name tetris_written stripes_written partial_stripes full_stripes blocks_written blocks_read raid_read_io_latenc y_histo raid_tetris_latency_ histo Size 8B 8B 8B 8B 8B 8B 8B 8B Type RATE RATE RATE RATE RATE RATE Display Unit PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC Description Tetrises written per second Stripes written per second Partial stripes written per second Full stripes written per second Blocks written per second Blocks read per second Raid Read I/O Latency Histogram Raid Tetris Latency Histogram Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

DELTA NONE DELTA NONE

Performance Management Design Guide

99 of 138

3/7/2008

raid_stripe object Counter Name Size Type stripe_size 8B RAW Display Unit NONE Description Size of this stripe Priority CM_DIAG CM_DIAG Depends on

Histogram of blocks per stripe_histogram 8B RATE PER_SEC stripe written per second for a given stripe size

replication object Counter Name rsm_rw_ent_allocs Size 8B Type RAW Display Unit NONE Description Total number of rsm_rw_ents resources allocated in semi-sync mode Total number of rsm_rw_ents resources free'd in semi-sync mode Priority CM_DIAG Depends on

rsm_rw_ent_frees src_throughput dst_throughput

8B 8B 8B

RAW RATE RATE RAW

NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

KB_PER_S Total replication source EC throughput KB_PER_S Total replication EC destination throughput SECS Concurrent transfers duration for source transfers Concurrent transfers duration for destination transfers Outstanding RAID/WAFL messages issued by VSM Total number of stolen buffers for subsystems Total number of unstolen buffers for subsystems Total number of inodes processed by SnapDiff

src_concurrent_transf 8B ers dst_concurrent_transf 8B ers vsm_async_msgs_in _use 8B

RAW

SECS

CM_DIAG

RAW RAW RAW RAW

NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

bkp_rcv_buf_counts_ 8B stolen bkp_rcv_buf_counts_ 8B unstolen sd_num_inodes_done 8B

Performance Management Design Guide

100 of 138

3/7/2008

sd_num_inodefile_bl ocks_done sd_num_files_done sd_num_file_blocks_ done sd_num_dirs_done sd_num_dir_blocks_ done

8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE

Total number of inodefile blocks processed by SnapDiff Total number of files processed by SnapDiff Total number of file blocks processed by SnapDiff Total number of directory processed by SnapDiff Total number of indoes processed by SnapDiff

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

rsm_src_global object Counter Name Size Type Display Unit Description Priority CM_DIAG Depends on

Numuber of RSM pre-allocated outstanding_waflbuf 4B RAW NONE WAFL buffers being used for CP sync at RSM source outstanding_mbuf Numuber of RSM pre-allocated 4B RAW NONE mbufs being used for CP sync at RSM source Numuber of in-flight volio 4B RAW NONE messages for CP sync at RSM source Numuber of in-flight plexio 4B RAW NONE messages for CP sync at RSM source Numuber of suspends of RSM_PLEXIO_START message 8B RAW NONE due to unavailable RSM preallocated WAFL buffers at RSM source Numuber of suspends of RSM_PLEXIO_START message 8B RAW NONE due to unavailable RSM preallocated mbufs at RSM source 8B RAW NONE Numuber of suspends of RSM_VOLIO_START message due

CM_DIAG

outstanding_volio

CM_DIAG

outstanding_plexio

CM_DIAG

suspend_waflbuf

CM_DIAG

suspend_mbuf

CM_DIAG

suspend_plexio

CM_DIAG

Performance Management Design Guide

101 of 138

3/7/2008

to unavailable RSM pre-allocated plexio messages at RSM source

rsm_dst_global object Counter Name outstanding_logrec v_msg outstanding_wafl_ msg outstanding_worki o outstanding_netad min_msg Size Type Display Unit NONE Description Number of in-flight logrecv messages for NVLOG sync at RSM destination Number of in-flight wafl messages for NVLOG sync at RSM destination Number of in-flight workio messages for CP sync at RSM destination Priority Depends on

4B

RAW

CM_DIAG

4B

RAW

NONE

CM_DIAG

4B

RAW

NONE

CM_DIAG

4B

RAW

NONE

Number of netadmin messages un-acked by CM_DIAG rsm_netadmin thread for CP sync at RSM destination Number of suspends due to unavailable workio messages at RSM destination CM_DIAG

suspend_workio

8B

RAW

NONE

target object Counter Name read_ops write_ops other_ops read_data write_data Size Type Display Unit Description Read operations per second Write operations per second Other operations per second Read bytes from filer per second Priority CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC Depends on

8B RATE PER_SEC 8B RATE PER_SEC 8B RATE PER_SEC 8B RATE B_PER_SEC

8B RATE B_PER_SEC Write bytes to filer per

Performance Management Design Guide

102 of 138

3/7/2008

second queue_full 8B RATE PER_SEC SCSI queue full responses per second CM_BASIC

lun object Counter Name display_name read_ops write_ops other_ops read_data write_data queue_full avg_latency total_ops scsi_partner_ops Size 64B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type STRING RATE RATE RATE RATE RATE RATE AVERA GE RATE RATE Display Unit NONE PER_SEC PER_SEC PER_SEC B_PER_SEC B_PER_SEC PER_SEC MSECS PER_SEC PER_SEC Description Name of the lun Read operations per second Write operations per second Other operations per second Read bytes per second Write bytes per second Queue full responses per second Average latency in milliseconds for all operations on the LUN Total number of operations on the LUN per second SCSI operations per second received from the partner node in a clustered system SCSI data in bytes per second read and written from the partner node in a clustered system Priority CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_DIAG total_ops Depends on

scsi_partner_data

8B

RATE

B_PER_SEC

CM_DIAG

read_align_histo

8B

PERCT

PERCT

Histogram of wafl read op CM_DIAG|CM alignment (number sectors off _ASUP_HOUR read_ops wafl block start) LY Histogram of wafl write op alignment (number of sectors off wafl block start) CM_DIAG|CM _ASUP_HOUR write_ops LY

write_align_histo read_partial_bloc ks read_ops1

8B

PERCT

PERCT

8B 8B

PERCT RATE NODISP

PERCT RATE NODISP

Percent of reads whose size is CM_DIAG|CM not a multiple of wafl block _ASUP_HOUR read_ops1 size LY Read operations per second CM_DIAG|CM _ASUP_HOUR

Performance Management Design Guide

103 of 138

3/7/2008

LY write_partial_bloc ks write_ops1 8B PERCT RATE NODISP PERCT RATE NODISP Percent of writes whose size CM_DIAG|CM is not a multiple of wafl block _ASUP_HOUR write_ops1 size LY Write operations per second CM_DIAG|CM _ASUP_HOUR LY

8B

fcp object Counter Name fcp_ops fcp_latency Size 8B 8B Type RATE AVERAGE Display Unit PER_SEC MSECS Description FCP operations per second Average latency for FCP operations in milliseconds Histogram of latency for FCP operations in milliseconds Priority CM_BASIC CM_BASIC fcp_ops Depends on

fcp_latency_hist fcp_write_data fcp_read_data

8B 8B 8B

DELTA RATE RATE

NONE

CM_DIAG CM_BASIC CM_BASIC

B_PER_SE FCP bytes written per C second B_PER_SE FCP bytes read per C second Average latency for read operations observed over all LUNs in the system accessed over FCP in milliseconds Histogram of latency for FCP read operations in milliseconds Total number of read operations per second observed over all the LUNS in the system accessed over FCP Average latency for write operations observed over all LUNs

fcp_read_latenc y

8B

AVERAGE

MSECS

CM_DIAG

fcp_read_ops

fcp_read_latenc y_hist

8B

DELTA

NONE

CM_DIAG

fcp_read_ops

8B

RATE

PER_SEC

CM_DIAG

fcp_write_latenc y

8B

AVERAGE

MSECS

CM_DIAG

fcp_write_ops

Performance Management Design Guide

104 of 138

3/7/2008

in the system accessed over FCP in milliseconds fcp_write_latenc 8B y_hist DELTA NONE Histogram of latency for FCP write operations in milliseconds Total number of write operations per second observed over all LUNs in the system accessed over FCP Average latency for read operations observed over all LUNs in the partner system accessed over FCP in milliseconds Total number of read operations per second observed over all the LUNS in the partner system accessed over FCP Average latency for write operations observed over all LUNs in the partner system accessed over FCP in milliseconds Total number of write operations per second observed over all LUNs in the partner system accessed over FCP Histogram of FCP read sizes Histogram of FCP write sizes CM_DIAG

fcp_write_ops

8B

RATE

PER_SEC

CM_DIAG

fcp_partner_rea d_latency

8B

AVERAGE

MSECS

CM_DIAG

fcp_partner_rea d_ops

fcp_partner_rea d_ops

8B

RATE

PER_SEC

CM_DIAG

fcp_partner_writ 8B e_latency

AVERAGE

MSECS

CM_DIAG

fcp_partner_wr ite_ops

fcp_partner_writ 8B e_ops fcp_read_size_h isto fcp_write_size_ histo

RATE

PER_SEC

CM_DIAG

4B 4B

DELTA DELTA

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

105 of 138

3/7/2008

iscsi object Counter Name iscsi_ops iscsi_latency Size 8B 8B Type RATE AVERA GE DELTA RATE RATE Display Unit PER_SEC MSECS Description iSCSI operations per second Average latency for ISCSI operations in milliseconds Histogram of latency for ISCSI operations in milliseconds Priority CM_BASIC CM_BASIC iscsi_ops Depends on

iscsi_latency_hist iscsi_write_data iscsi_read_data

8B 8B 8B

NONE

CM_DIAG CM_BASIC CM_BASIC

B_PER_S iSCSI bytes written EC per second B_PER_S iSCSI bytes read per EC second Average latency of read operations observed over all LUNs in the system accessed over iSCSI in milliseconds Histogram of latency for ISCSI read operations in milliseconds

iscsi_read_latency

8B

AVERA GE

MSECS

CM_DIAG

iscsi_read_ops

iscsi_read_latency_hi st

8B

DELTA

NONE

CM_DIAG

iscsi_read_ops

8B

RATE

Total number of read operations per second PER_SEC observed over all the LUNs in the system accessed over iSCSI Average latency of write operations observed over all LUNs in the system accessed over iSCSI in milliseconds Histogram of latency for ISCSI write operations in milliseconds Total number of write

CM_DIAG

iscsi_write_latency

8B

AVERA GE

MSECS

CM_DIAG

iscsi_write_op s

iscsi_write_latency _hist iscsi_write_ops

8B 8B

DELTA RATE

NONE PER_SEC

CM_DIAG CM_DIAG

Performance Management Design Guide

106 of 138

3/7/2008

operations per second observed over all the LUNs in the system accessed over iSCSI iscsi_read_size_histo iscsi_write_size_histo 4B 4B DELTA DELTA NONE NONE Histogram of ISCSI read sizes Histogram of ISCSI write sizes CM_DIAG CM_DIAG

sparse object Counter Name sent_ops sent_op_count sent_op_percent outstanding_ops outstanding_op_count failed_ops reclaimed_ops nrv_op_remote_latenc y response_rcvd_ops callback_rcvd_ops callback_rcvd_op_cou nt resp_xidmatch Size 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type DELTA DELTA PERCT RAW RAW DELTA DELTA AVERA GE DELTA DELTA DELTA DELTA Display Unit NONE NONE PERCT NONE NONE NONE NONE MSECS NONE NONE NONE NONE Description Total NRV ops sent from sparse volume NRV ops sent from sparse volume NRV ops sent percentage from sparse volume Priority CM_DIAG CM_DIAG CM_DIAG sent_ops Depends on

Total NRV ops outstanding CM_DIAG from sparse volume NRV ops outstanding from sparse volume Total NRV ops failed from sparse volume Total NRV ops reclaimed in sparse volume Average NRV op remote latency in sparse volume Total NRV ops received response in sparse volume Total NRV callback ops received for sparse volume NRV callback ops received for sparse volume NRV responses with matching xid in sparse volume CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG response_rcvd _ops

Performance Management Design Guide

107 of 138

3/7/2008

resp_xidnomatch

8B

DELTA

NONE

NRV responses with no matching xid in sparse volume NRV responses/callbacks dropped due to bad contents in sparse volume NRV responses/callbacks not processed due to no memory in sparse volume Active NRV connections in sparse volume Failed NRV connections in sparse volume Successful NRV connections in sparse volume Outstanding NRV connections in sparse volume Reclaimed NRV connections in sparse volume NRV connections closed by heartbeat module in sparse volume

CM_DIAG

resp_bad

8B

DELTA

NONE

CM_DIAG

resp_nomem conn_active conn_failure conn_success

8B 8B 8B 8B

DELTA RAW DELTA DELTA

NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

conn_outstanding

8B

RAW

NONE

CM_DIAG

conn_reclaimed

8B

DELTA

NONE

CM_DIAG

conn_htbeat_closed nrv_bytes_sent nrv_bytes_received read_xfers read_blocks readahead_xfers readahead_blocks sreq_nomem

8B 8B 8B 8B 8B 8B 8B 8B

DELTA RATE RATE DELTA DELTA DELTA DELTA DELTA

NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

B_PER_S Bytes sent from sparse EC volume B_PER_S Bytes received for sparse EC volume NONE NONE NONE NONE NONE NRV read transfers in sparse volume NRV read blocks in sparse volume NRV readahead transfers in sparse volume NRV readahead blocks in sparse volume Sparse requests denied due to no memory in sparse volume

Performance Management Design Guide

108 of 138

3/7/2008

sreq_enqueued sreq_dequeued sreq_done sreq_overdemand

8B 8B 8B 8B

DELTA DELTA DELTA DELTA

NONE NONE NONE NONE

Sparse requests enqueued in sparse volume Sparse requests dequeued to active state in sparse volume Sparse requests completed in sparse volume Sparse requests denied due to exceeding demand limit in sparse volume Sparse requests denied due to excedding speculative read-ahead limit in sparse volume Sparse requests currently waiting to be dequeued in sparse volume Sparse requests currently active in sparse volume Sparse requests currently free in sparse volume Sparse coalesce lookups in sparse volume Sparse coalesce comparisons in sparse volume Sparse coalesce hits in sparse volume Sparse coalesce partials in sparse volume Number of sparse processe for sparse volumes Number of sparse process allocations that failed due to no memor for sparse volumey Number of sparse processes exited for sparse volume

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

sreq_overspecra

8B

DELTA

NONE

CM_DIAG

sreq_curr_waiting sreq_curr_active sreq_curr_free coalesce_lookups coalesce_comparisons coalesce_hits coalesce_partials sreq_pump_curr_npro c sreq_pump_nomem

8B 8B 8B 8B 8B 8B 8B 8B

DELTA DELTA DELTA DELTA DELTA DELTA DELTA DELTA

NONE NONE NONE NONE NONE NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

8B

DELTA

NONE

CM_DIAG

sreq_exit_nproc

8B

DELTA

NONE

CM_DIAG

Performance Management Design Guide

109 of 138

3/7/2008

spinhi object Counter Name spinhi_fileops spinhi_filecbs Size 8B 8B Type RATE RATE AVERA GE DELTA NODIS P RATE AVERA GE DELTA NODIS P RATE DELTA DELTA DELTA NODIS P DELTA NODIS P DELTA NODIS P DELTA NODIS P Display Unit PER_SEC Description Total number of spinhi file operations per second Priority CM_BASIC CM_ADMI N CM_BASIC Depends on

Total number of spinhi PER_SEC callback operations per second USECS DELTA NODISP PER_SEC USECS DELTA NODISP PER_SEC NONE NONE DELTA NODISP DELTA NODISP DELTA NODISP DELTA NODISP Average latency for spinhi read operations in microseconds Number of spinhi reads for latency calculation Total observed spinhi read operations per second Average latency for spinhi write operations in microseconds Number of spinhi writes for latency calculation

spinhi_read_latency spinhi_avg_read_late ncy_base spinhi_read_ops spinhi_write_latency

8B

spinhi_avg_read_latenc

8B 8B 8B

CM_BASIC CM_BASIC CM_BASIC

spinhi_avg_write_laten

spinhi_avg_write_late 8B ncy_base spinhi_write_ops spinhi_fileop_count spinhi_filecb_count spinhi_fileop_latency _base spinhi_fileop_cputim e_base spinhi_filecb_latency _base 8B 8B 8B 8B

CM_BASIC

Total observed spinhi write CM_BASIC operations per second Array of select spinhi file operation counts Array of select spinhi callback operation counts Array of select spinhi file operation counts for latency calculations Array of select spinhi file operation counts for cputime calculations Array of select spinhi callback operation counts for latency calculations Array of select spinhi callback operation counts for cputime calculations CM_DIAG CM_DIAG CM_DIAG

8B

CM_DIAG

8B

CM_DIAG

spinhi_filecb_cputime 8B _base

CM_DIAG

Performance Management Design Guide

110 of 138

3/7/2008

spinhi_fileop_latency spinhi_fileop_cputim e spinhi_filecb_latency

8B 8B 8B

AVERA GE AVERA GE AVERA GE AVERA GE AVERA GE DELTA NODIS P AVERA GE DELTA NODIS P AVERA GE DELTA NODIS P AVERA GE DELTA NODIS P RAW RAW RATE RATE

USECS USECS USECS USECS USECS DELTA NODISP USECS DELTA NODISP USECS DELTA NODISP USECS DELTA NODISP NONE NONE

Array of latencies of select spinhi file operations Array of cputimes of select spinhi file operations Array of latencies of select spinhi callback operations Array of cputimes of select spinhi callback operations Average latency of the spinhi file ops Number of spinhi file ops for average latency calculation Average cputime of the spinhi file ops Number of spinhi file ops for average cputime calculation Average latency of the spinhi callback ops Number of spinhi callback ops for average latency calculation Average cputime of the spinhi callback ops Number of spinhi callback ops for average cputime calculation Number of spinhi file operations in progress Number of spinhi callback operations in progress

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

spinhi_fileop_latency_b

spinhi_fileop_cputime_

spinhi_filecb_latency_b

spinhi_filecb_cputime 8B spinhi_avg_fileop_lat ency spinhi_avg_fileop_lat ency_base spinhi_avg_fileop_cp utime spinhi_avg_fileop_cp utime_base spinhi_avg_filecb_lat ency spinhi_avg_filecb_lat ency_base spinhi_avg_filecb_cp utime spinhi_avg_filecb_cp utime_base spinhi_fileops_curren t spinhi_filecb_current spinhi_read_kbytes spinhi_write_kbytes 8B 8B 8B 8B 8B 8B 8B 8B 4B 4B 8B 8B

spinhi_filecb_cputime_

spinhi_avg_fileop_laten

spinhi_avg_fileop_cput

spinhi_avg_filecb_laten

spinhi_avg_filecb_cput

KB_PER_ Number of kbytes read by SEC clients KB_PER_ Number of kbytes written SEC by clients

Performance Management Design Guide

111 of 138

3/7/2008

system object Counter Name system_model ontap_version serial_no system_id hostname nfs_ops cifs_ops http_ops fcp_ops iscsi_ops read_ops sys_read_latency sys_read_latency_hist write_ops sys_write_latency Size 16B 252B 16B 64B 64B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type STRING STRING STRING STRING STRING RATE RATE RATE RATE RATE RATE AVERAGE DELTA RATE AVERAGE Display Unit NONE NONE NONE NONE NONE PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC MSECS NONE PER_SEC MSECS Description Name of the system model ONTAP version System serial number System ID Hostname NFS operations per second CIFS operations per second HTTP operations per second FCP operations per second iSCSI operations per second Read operations per second Average latency for all read operations in the system in milliseconds Histogram of latency for all read operations in milliseconds Write operations per second Average latency for all write operations in the system in milliseconds Histogram of latency for all write operations in milliseconds Total operations per second Average latency for all write operations in the system in milliseconds Histogram of latency for all operations in milliseconds Network KB received per second Network KB sent per second Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC CM_BASIC CM_BASIC&~CM _GX CM_BASIC&~CM _GX CM_BASIC&~CM _GX CM_BASIC CM_DIAG CM_DIAG CM_BASIC CM_DIAG

sys_write_latency_hist total_ops sys_avg_latency sys_latency_hist net_data_recv net_data_sent

8B 8B 8B 8B 8B 8B

DELTA RATE AVERAGE DELTA RATE RATE

NONE PER_SEC MSECS NONE KB_PER_SE C KB_PER_SE C

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_BASIC&~CM _GX CM_BASIC&~CM _GX

Performance Management Design Guide

112 of 138

3/7/2008

disk_data_read disk_data_written cpu_busy cpu_elapsed_time avg_processor_busy cpu_elapsed_time1 total_processor_busy cpu_elapsed_time2 num_processors time uptime

8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B

RATE RATE PERCT DELTA NODISP PERCT DELTA NODISP PERCT DELTA NODISP RAW RAW RAW

KB_PER_SE C KB_PER_SE C PERCT DELTA NODISP PERCT DELTA NODISP PERCT DELTA NODISP NONE SECS SECS

Disk KB read per second Disk KB written per second System CPU resource utilization Elapsed time since boot Average processor utilization across all processors in the system Elapsed time since boot Total processor utilization of all processors in the system Elapsed time since boot Number of active processors in the system

CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_ADMIN CM_BASIC CM_ADMIN CM_BASIC CM_ADMIN

Time in seconds since the CM_STATS|CM_T Epoch (00:00:00 UTCJanuary EST 11970) Time in seconds that the system has been up CM_STATS|CM_T EST

Performance Management Design Guide

113 of 138

3/7/2008

perf object Counter Name ops_per_sec_histo data_disk_util_percnt_histo parity_disk_util_percnt_histo disk_read_MB_per_sec_histo disk_write_MB_per_sec_histo netin_per_sec_histo Size Type Display Unit Description Histogram of operations per second handled by the appliance Histogram of data disk utilization percentage Histogram of parity disk utilization percentage Histogram of disk read megabytes per second Histogram of disk write megabytes per second Histogram of network data per second coming in to the appliance Histogram of network data per second going out of the appliance Matrix of per second correlation between CPU utilization and disk utilization Matrix of CSMP domain switches domain utilization Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG Depends on

8B DELTA NONE 8B DELTA NONE 8B DELTA NONE 8B DELTA NONE 8B DELTA NONE 8B DELTA NONE

netout_per_sec_histo

8B DELTA NONE

CM_DIAG CM_DIAG|C M_ASUP_HO URLY CM_DIAG CM_DIAG|C M_ASUP_HO URLY

cpu_disk_util_matrix

8B DELTA NONE

domain_switches domain_usecs

4B RATE

PER_SEC

8B DELTA USECS

shared_domain_usecs

8B DELTA USECS

CM_DIAG|C shared domain utilization M_ASUP_HO URLY CP interrupt utilization non-CP interrupt utilization Disk CP reads latency histogram Disk user reads latency histogram Disk user writes latency histogram CM_DIAG|C M_ASUP_HO URLY CM_DIAG|C M_ASUP_HO URLY CM_DIAG CM_DIAG CM_DIAG

cp_rupt_usecs

8B DELTA USECS

noncp_rupt_usecs disk_cp_reads_latency_histogram disk_user_reads_latency_histogram

8B DELTA USECS 8B DELTA NONE 8B DELTA NONE 8B DELTA NONE

disk_user_writes_latency_histogra m

Performance Management Design Guide

114 of 138

3/7/2008

tape object Counter Name aliases bytes_read bytes_written write_commands _hist Size 252B 8B 8B Type STRIN G DELTA DELTA Display Unit NONE NONE NONE Description Alias names of the tape drive Total number of bytes read from tape Total number of bytes written to tape Array of number of write commands of various record size ranges Priority CM_DIAG CM_DIAG CM_DIAG Depends on

8B

DELTA

NONE

CM_DIAG

write_rate_hist

8B

Array of average data AVERA KB_PER_SE rates of writes of GE C various record size ranges AVERA GE DELTA NODIS P MSECS Array of average latencies of writes of various record size ranges Array of command execution times of writes of various record size ranges Array of number of read commands of various record size ranges

CM_DIAG

write_time_hist

write_latency_his t

8B

CM_DIAG

write_commands_ hist

write_time_hist

8B

DELTA NODISP

CM_DIAG

read_commands_ hist

8B

DELTA

NONE

CM_DIAG

read_rate_hist

8B

Array of average data AVERA KB_PER_SE rates of reads of GE C various record size ranges AVERA GE DELTA NODIS MSECS DELTA NODISP Array of average latencies of reads of various record size ranges Array of command execution times for

CM_DIAG

read_time_hist

read_latency_hist

8B

CM_DIAG

read_commands_hi st

read_time_hist

8B

CM_DIAG

Performance Management Design Guide

115 of 138

3/7/2008

reads of various record size ranges

test object Counter Name counter_0 counter_1 counter_2 counter_3 counter_4 counter_5 counter_6 counter_7 counter_8 counter_9 counter_10 counter_11 Size Type Display Unit Description Priority CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST CM_TEST Depends on

4B RAW 8B RAW 4B RAW 8B RAW 4B RAW 8B RAW 4B RAW 8B RAW 4B RATE 4B RATE

PER_SEC 4 bytehex valueraw PER_SEC 8 bytehex valueraw PER_SEC 4 bytehex valueraw PER_SEC 8 bytehex valueraw PER_SEC 4 bytedecimal valueraw PER_SEC 8 bytedecimal valueraw PER_SEC 4 bytedecimal valueraw PER_SEC 8 bytedecimal valueraw PER_SEC 4 bytehex valuerate PER_SEC 4 bytedecimal valuerate NONE Test Histogramhex value

4B DELTA PER_SEC 4 bytehex valuedelta 4B DELTA PER_SEC 4 bytedecimal valuedelta

test_histo_0 4B RAW

vfiler object Counter Name vfiler_cpu_busy vfiler_cpu_busy _base vfiler_net_data_r ecv Size 8B Type PERCT DELTA NODIS P RATE RATE Display Unit PERCT DELTA NODISP Description Percentage CPU time used by the vfiler Base for percentage CPU time used by the vfiler Priority CM_BASIC Depends on vfiler_cpu_busy_base

8B

CM_BASIC

8B

Network KB KB_PER_SEC received per second by the vfiler KB_PER_SEC Network KB sent per second by the

CM_BASIC CM_BASIC

vfiler_net_data_s 8B ent

Performance Management Design Guide

116 of 138

3/7/2008

vfiler vfiler_read_ops vfiler_write_ops vfiler_misc_ops vfiler_read_byte s vfiler_write_byte s 8B 8B 8B 8B 8B RATE RATE RATE RATE RATE PER_SEC PER_SEC PER_SEC KB_PER_SEC KB_PER_SEC Number of read ops for the vfiler Number of write ops for the vfiler Number of other miscellaneous ops for the vfiler Number of bytes read by the vfiler Number of bytes written by the vfiler CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC

volume object Counter Name parent_aggr avg_latency total_ops read_data read_latency read_ops write_data write_latency write_ops other_latency other_ops internal_msgs Size 64B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type STRING AVERAGE RATE RATE AVERAGE RATE RATE AVERAGE RATE AVERAGE RATE RATE Display Unit NONE USECS PER_SEC Description Name of hosting aggregate Average latency in microseconds for all operations on the volume Number of operations per second serviced by the volume Priority CM_DIAG CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_BASIC CM_DIAG

B_PER_SE Bytes read per second from the C volume USECS PER_SEC Average time for reads to the volume Number of reads per second to the volume

B_PER_SE Bytes written per second to the C volume USECS PER_SEC USECS PER_SEC PER_SEC Average time for writes to the volume Number of writes per second to the volume Average time for other operations to the volume Number of other operations per second to the volume Number of internal backdoor

Performance Management Design Guide

117 of 138

3/7/2008

messages per second to the volume read_blocks write_blocks synchronous_frees asynchronous_frees df_worker nfs_read_data nfs_read_latency nfs_read_ops nfs_write_data nfs_write_latency nfs_write_ops nfs_other_latency nfs_other_ops cifs_read_data cifs_read_latency cifs_read_ops cifs_write_data cifs_write_latency cifs_write_ops 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B RATE RATE RATE RATE RATE RATE AVERAGE RATE RATE AVERAGE RATE AVERAGE RATE RATE AVERAGE RATE RATE AVERAGE RATE PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC Number of blocks read per second from the volume Number of blocks written per second to the volume Number of synchronous frees per second to the volume Number of asynchronous frees per second to the volume Number of times per second delayed free worker called on the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X CM_DIAG CM_DIAG CM_DIAG CM_ADVANCED CM_ADVANCED CM_ADVANCED

B_PER_SE Bytes read per second via NFS C from the volume USECS PER_SEC Average time for NFS reads to the volume Number of NFS reads per second to the volume

B_PER_SE Bytes written per second via NFS C to the volume USECS PER_SEC USECS PER_SEC Average time for NFS writes to the volume Number of NFS writes per second to the volume Average time for other NFS operations to the volume Number of other NFS operations per second to the volume

CM_DIAG&~CM_G X CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED CM_ADVANCED

B_PER_SE Bytes read per second via cifs from C the volume USECS PER_SEC Average time for cifs reads to the volume Number of cifs reads per second to the volume

B_PER_SE Bytes written per second via cifs to C the volume USECS PER_SEC Average time for cifs writes to the volume Number of cifs writes per second to

Performance Management Design Guide

118 of 138

3/7/2008

the volume cifs_other_latency cifs_other_ops san_read_data san_read_latency san_read_ops san_write_data san_write_latency san_write_ops san_other_latency san_other_ops fcp_read_data fcp_read_latency fcp_read_ops fcp_write_data fcp_write_latency fcp_write_ops fcp_other_latency fcp_other_ops iscsi_read_data 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B AVERAGE RATE RATE AVERAGE RATE RATE AVERAGE RATE AVERAGE RATE RATE AVERAGE RATE RATE AVERAGE RATE AVERAGE RATE RATE USECS PER_SEC Average time for other cifs operations to the volume Number of other cifs operations per second to the volume CM_ADVANCED CM_ADVANCED

B_PER_SE Bytes read per second via block C protocol from the volume MSECS PER_SEC Average time for block protocol reads to the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

Number of block protocol reads per CM_DIAG&~CM_G X second to the volume

B_PER_SE Bytes written per second via block C protocol to the volume MSECS PER_SEC MSECS PER_SEC Average time for block protocol writes to the volume Number of block protocol writes per second to the volume Average time for other block protocol operations to the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

Number of other block protocol CM_DIAG&~CM_G operations per second to the volume X

B_PER_SE Bytes read per second via block C protocol from the volume MSECS PER_SEC Average time for block protocol reads to the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

Number of block protocol reads per CM_DIAG&~CM_G X second to the volume

B_PER_SE Bytes written per second via block C protocol to the volume MSECS PER_SEC MSECS PER_SEC Average time for block protocol writes to the volume Number of block protocol writes per second to the volume Average time for other block protocol operations to the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

Number of other block protocol CM_DIAG&~CM_G operations per second to the volume X

B_PER_SE Bytes read per second via block C protocol from the volume

CM_DIAG&~CM_G X

Performance Management Design Guide

119 of 138

3/7/2008

iscsi_read_latency iscsi_read_ops iscsi_write_data iscsi_write_latency iscsi_write_ops iscsi_other_latency iscsi_other_ops flexcache_read_dat a flexcache_read_ops flexcache_write_dat a flexcache_write_op s flexcache_other_op s flexcache_send_dat a flexcache_receive_ data wv_fsid wv_vol_type wv_fsinfo_fs_versi on wv_volinfo_fs_opti ons wv_volinfo_fs_flag s wv_fsinfo_blks_tot al

8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 16B 16B 4B 16B 16B 8B

AVERAGE RATE RATE AVERAGE RATE AVERAGE RATE RATE RATE RATE RATE RATE RATE RATE STRING STRING RAW STRING STRING RAW

MSECS PER_SEC

Average time for block protocol reads to the volume

CM_DIAG&~CM_G X

Number of block protocol reads per CM_DIAG&~CM_G second to the volume X

B_PER_SE Bytes written per second via block C protocol to the volume MSECS PER_SEC MSECS PER_SEC Average time for block protocol writes to the volume Number of block protocol writes per second to the volume Average time for other block protocol operations to the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

Number of other block protocol CM_DIAG&~CM_G operations per second to the volume X

B_PER_SE Bytes read per second via C FlexCache from the volume PER_SEC Number of FlexCache read operations per second from the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X

B_PER_SE Bytes written per second via C FlexCache to the volume PER_SEC PER_SEC

CM_DIAG&~CM_G X

Number of FlexCache write CM_DIAG&~CM_G operations per second to the volume X

Number of other FlexCache CM_DIAG&~CM_G operations per second to the volume X

B_PER_SE Bytes sent per second via C FlexCache from the volume B_PER_SE Bytes received per second via C FlexCache to the volume NONE NONE NONE NONE NONE NONE File system ID for the volume Volume type for the volume File system version for the volume File system options for the volume File system flags for the volume Total blocks in the volume

CM_DIAG&~CM_G X

CM_DIAG&~CM_G X CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

120 of 138

3/7/2008

wv_fsinfo_blks_res erve wv_fsinfo_blks_use d wv_fsinfo_blks_use d_by_plane0 wv_fsinfo_blks_blk s_rsrv_holes_cifs wv_fsinfo_blks_blk s_rsrv_holes wv_fsinfo_blks_blk s_rsrv_overwrite wv_fsinfo_blks_rsr v_absents wv_fsinfo_blks_sna p_reserve_pct wv_fsinfo_blks_res _state

8B 8B 8B 8B 8B 8B 8B 4B 4B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE PERCT NONE PERCT NONE NONE NONE NONE NONE NONE NONE NONE

Reserved blocks in the volume Used blocks (all planes) in the volume Used blocks (plane 0) in the volume Reserved blocks for CIFS holes in the volume Reserved blocks for all holes in the volume Overwrite reserved blocks in the volume Absent sparse volume reserved blocks in the volume Snap reserved percentage for the volume Space reservation state for the volume Overwrite slider percentage for the volume Total inodes in the volume Reserved inodes in the volume Used inodes in the volume CP when block reallocation first performed Blocks past EOF in the volume Snap reserved blocks in the volume Blocks used in zombie files in the volume Partial space reservation blocks charged to the aggregate for the volume Whether space reservation charges are made to the aggregate for the volume

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

wv_fsinfo_blks_ove 4B rwrite_slider_pct wv_fsinfo_inos_tot al wv_fsinfo_inos_res erve wv_fsinfo_inos_use d wv_fsinfo_blkr_cp wvblk_past_eof wvblk_snap_reserv e wvblk_zombie_blks wv_fsinfo_blks_rsr v_parent wvblk_spcres_in_p arent 4B 4B 4B 4B 8B 8B 8B 8B

4B

RAW

NONE

CM_DIAG

Performance Management Design Guide

121 of 138

3/7/2008

wvblk_rsrv_parent_ holes wvblk_rsrv_parent_ overwrite wvblk_rsrv_parent_ overwrite_always wvbd_active_frees wvbd_whole_frees wvbd_active_frees_ y wvbd_owner_chang ed_y wvbd_whole_frees_ o wvblk_saved_fsinfo _inos_total wvblk_saved_fsinfo _inos_reserve wvblk_saved_fsinfo _inos_used wvblk_delalloc wvblk_rsrv_delallo c wvblk_rsrv_holes_c ifs wvblk_rsrv_holes

8B 8B 8B 8B 8B 8B 8B 8B 4B 4B 4B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

Blocks charged to the aggregate for holes in the volume Blocks charged to the aggregate for overwrites in the volume Blocks charged to the aggregate for always overwrites in the volume Active file system freed blocks during CP in the volume Active file system whole freed blocks during CP in the volume Active file system freed blocks (youngest) during CP in the volume Changed ownership blocks (youngest) during CP in the volume Whole freed blocks (oldest) during CP in the volume Total inodes during CP in the volume Reserved inodes during CP in the volume Used inodes during CP in the volume Delalloc blocks during CP in the volume Reserved delalloc blocks during CP in the volume Blocks reserved for CIFS holes during CP in the volume Blocks reserved for all holes during CP in the volume Blocks reserved for overwrite during CP in the volume Absent sparse volume reserved blocks during CP in the volume Extent size Blocks from a clone volume associated with the parent volume Delayed frees are enabled for this volume

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

wvblk_rsrv_overwri 8B te wvblk_rsrv_absents extent_size wvblk_rsrv_clone_ parent wvdf_enabled 8B 4B 8B 4B

Performance Management Design Guide

122 of 138

3/7/2008

wvdf_last_fbn wvdf_num_frees wvdf_water_mark wvdf_max_frees wvdf_total_score wvblk_parent_to_b e_reclaimed

8B 8B 8B 8B 4B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RATE RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

Last active map block from which we freed a pvbn Number of delayed pvbn frees from the flex volume pvbn frees start when wvdf_num_frees passes this limit Maximum number of delayed pvbn frees Total delayed free score across entire active map Outstanding delayed pvbn freesindependent of score cap in the flex volume Container file fileid Container blocks in the flex volume Container size Indirect blocks in the container file of the flex volumes Average number of requests suspended on volume For a clone volume an estimate of the storage directly used by clone Last start time of the block reclamation scanner Last reset time of the block reclamation scanner Last completion time of the block reclamation scanner Last abort time of the block reclamation scanner Number of zombie/sfsr messages working Delalloc blocks during CP in the volume Last FBN till which indirects have been counted

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_TEST CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

wvip_vvol_containe 4B r_wi_fileid wvip_vvol_containe 8B r_wi_blk_cnt wvip_vvol_containe 8B r_wi_size wvip_vvol_containe 8B r_indirects wvol_number_susp ended clone_storage_bloc ks wvblk_reclaim_tim e_start wvblk_reclaim_tim e_reset wvblk_reclaim_tim e_done wvblk_reclaim_tim e_abort wvzmb_num_zmsg s_inuse wvblk_ind_delalloc wvsblk_vvrd_last_f bn 4B 8B 8B 8B 8B 8B 4B 8B 4B

Performance Management Design Guide

123 of 138

3/7/2008

wvsblk_vvrd_spcfla gs wvsblk_vvrd_spc_c lone_inherited wvsblk_space_tax wvsblk_lev0_over_ nominal

4B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

Space management scanner flags Inherited block count for the clone Space management tax for volume Count of level0 container blocks over nominal size

CM_DIAG CM_DIAG CM_DIAG CM_DIAG

wvsblk_vvrd_vol_si 4B ze wvsblk_vvrd_flags wv_fsinfo_contain ment_version_slc wv_fsinfo_contain ment_version_sle wv_fsinfo_contain ment_version_spare1 wv_fsinfo_contain ment_version_spare2 wv_fsinfo_contain ment_version_highes t_slc wv_fsinfo_contain ment_version_highes t_sle wv_fsinfo_contain ment_version_highes t_spare1 wv_fsinfo_contain ment_version_highes t_spare2 4B 4B 4B 4B 4B 4B

Nominal size stored in the raid label CM_DIAG Volume flags stored in the raid label Containment version for SnapLock Compliance volume Containment version for SnapLock Enterprise volume Containment version reserved for SnapLock Containment version reserved for SnapLock Highest containment version detected for SnapLock Compliance volume Highest containment version detected for SnapLock Enterprise volume Highest containment version reserved for SnapLock Highest containment version reserved for SnapLock CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

4B

RAW

NONE

CM_DIAG

4B

RAW

NONE

CM_DIAG

4B

RAW

NONE

CM_DIAG

volume_snapmirror_destination object Counter Name num_async_xfer Size Type Display Unit NONE Description Priority Depends on Number of asynchronous transfers initiated since CM_DIAG reboot

4B RAW

Performance Management Design Guide

124 of 138

3/7/2008

num_succ_async_xfer

4B RAW

NONE

Number of successful asynchronous transfers since reboot Number of restarts in this relationship since reboot Total number of blocks received in async transfer mode since reboot Number of raid(trad)/wafl(flex) read requests for current (or last) async transfer Number of block write requests to raid(trad)/wafl(flex) for current (or last) async transfer Number of blocks written to raid(trad)/wafl(flex) for current (or last) async transfer Slice number of the latest restart checkpoint Number of verify log blocks sent during fix opeation Number of verify transfers initiated since reboot Number of successful verify transfers since reboot Number of raid(trad)/wafl(flex) read requests for current (or last) verify transfer Number of block read requests to raid(trad)/wafl(flex) for current (or last) verify

CM_DIAG

num_restarts

4B RAW

NONE

CM_DIAG

total_blocks_recvd

4B RAW

NONE

CM_DIAG

blocks_recvd

4B RAW

NONE

CM_DIAG

blocks_write_request

4B RAW

NONE

CM_DIAG

blocks_write_done

4B RAW

NONE

CM_DIAG

checkpoint_slice fix_logblks_sent

4B RAW 4B RAW

NONE NONE

CM_DIAG CM_DIAG

vrf_num_xfer

4B RAW

NONE

CM_DIAG

vrf_num_succ_xfer

4B RAW

NONE

CM_DIAG

vrf_blocks_recvd

4B RAW

NONE

CM_DIAG

vrf_blocks_read_request 4B RAW

NONE

CM_DIAG

Performance Management Design Guide

125 of 138

3/7/2008

transfer vrf_blocks_read_done 4B RAW NONE Number of blocks read from raid(trad)/wafl(flex) for current (or last) verify transfer Number of mismatches for current (or last) verify transfer CM_DIAG

vrf_mismatches_found

4B RAW

NONE

CM_DIAG

vrf_mismatches_logged

4B RAW

NONE

Number of mismatches logged for current (or last) CM_DIAG verify transfer Number of indirect blocks logged for current (or last) verify transfer Number of late writes received Number of blocks in a write sent to WAFL/RAID Average number of blocks in a write req Number of writes sent to WAFL/RAID Histogram of time taken for a write request to complete Number of message suspensions during a write request Rate of message suspensions during a write request Number of times suspended during async message allocations Histogram of time taken for jumpahead CM_DIAG CM_DIAG CM_DIAG CM_DIAG write_tetris_cnt CM_DIAG CM_DIAG

vrf_indir_blocks late_writes write_tetris_size avg_write_tetris_size write_tetris_cnt write_req_time

4B RAW 8B RAW 8B RAW

NONE NONE NONE

8B AVERAGE NONE 8B RAW 8B RAW NONE NONE

num_write_req_susp

8B RAW

NONE

CM_DIAG

write_req_susp_rate

8B RATE

PER_SEC

CM_DIAG

num_async_msg_susp jumpahead_time

8B RAW 8B RAW

NONE NONE

CM_DIAG CM_DIAG

Performance Management Design Guide

126 of 138

3/7/2008

volume_snapmirror_source object Counter Name num_async_xfer num_succ_async_xfer num_restarts total_blocks_sent Size 4B 4B 4B 8B Type RAW RAW RAW RAW Display Unit NONE NONE NONE NONE Description Number of asynchronous transfers initiated since reboot Number of successful asynchronous transfers since reboot Number of restarts in this relationship since reboot Total number of blocks sent in async transfer mode since reboot Number of raid(trad)/wafl(flex) read requests for current (or last) async transfer Number of blocks read from raid(trad)/wafl(flex) for current (or last) async transfer Number of blocks sent over network for current (or last) async transfer Slice of current (or last) asynchronous transfer number of snapmap blocks examined by inode differ Time spent in inode differ Time bmb spent waiting for bufs to free up number of flush messages received by idmp_proc number of restore messages received by idmp_proc Time idmp_proc spent waiting for flush number of times flush done by idmp_proc Number of verify logblks received during fix operation 127 of 138 Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG

blocks_read_request

8B

RAW

NONE

CM_DIAG

blocks_read_done

8B

RAW

NONE

CM_DIAG

blocks_sent slice_in_progress idiff_num_snapmap_blk s_examined idiff_run_time bmb_wait_time_for_buf s_to_free num_flush_messages_re cvd num_restore_messages_ recvd idmp_proc_flush_wait_t ime num_flush_done fix_logblks_recvd

8B 8B 8B 8B 8B 8B 8B 8B 8B 4B

RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

NONE NONE NONE USECS USECS NONE NONE USECS NONE NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

3/7/2008

vrf_num_xfer vrf_num_succ_xfer

4B 4B

RAW RAW

NONE NONE

Number of verify transfers initiated since reboot Number of successful verify transfers since reboot Number of raid(trad)/wafl(flex) read requests for current (or last) async transfer Number of blocks read from raid(trad)/wafl(flex) for current verify transfer Number of blocks sent over network for current verify transfer Slice of current verify transfer

CM_DIAG CM_DIAG

vrf_blocks_read_request 8B

RAW

NONE

CM_DIAG

vrf_blocks_read_done

8B

RAW

NONE

CM_DIAG

vrf_blocks_sent vrf_slice_in_progress activemap_bkdoor_time

8B 8B 8B

RAW RAW RAW AVERAG E RAW RAW AVERAG E RAW RAW AVERAG E RAW RAW AVERAG E RAW

NONE NONE USECS USECS NONE USECS USECS NONE USECS USECS NONE NONE NONE MSECS

CM_DIAG CM_DIAG

Total time in backdoor calls for CM_DIAG activemap blks Average activemap block backdoor time Number of activemap block backdoor calls Total number of usecs taken to read a kireeti block Average kireeti block read time Number of kireeti blocks read for this transfer Total number of usecs taken to send blocks to network Average network block write time Number of blocks sent to network CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

avg_activemap_bkdoor_ 8B time num_activemap_bkdoor s kireeti_blk_time avg_kireeti_time num_kireeti_blk blk_write_time avg_blk_write_time blk_write_cnt read_req_len avg_read_req_len read_req_time_ctr 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B

num_act

num_kir

blk_writ

Number of data blocks read per WAFL/RAID request since CM_DIAG beginning Average number of data blocks per read request Total time taken for a chain of data blocks to be read 128 of 138 CM_DIAG CM_DIAG

read_req

Performance Management Design Guide

3/7/2008

avg_read_req_time_ctr read_req_cnt num_async_msg_waits cached_answer_hits cached_answer_misses read_req_time activemap_susp kireeti_susp num_wafl_req_susp num_blocked_q1_full num_blocked_q2_full cpu_bmb_time cpu_pipe_time incr_snaps

8B 8B 8B 8B 8B 8B 8B 8B 8B 4B 4B 8B 8B 8B

AVERAG E RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW RAW

MSECS NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE MSECS MSECS NONE

Avg time taken for a chain of data blocks to be read Number of WAFL/RAID data read requests Number of times suspended during async message allocations Number of hits in answers stored in cache Number of misses in answers stored in cache Histogram of time (msecs) taken for a chain of data blocks to be read

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

read_req

Number of times wafl CM_DIAG suspends on activemap blk reads Number of times wafl suspends on kireeti blk reads Histogram of number of message suspensions during a read request Number of times idmp_recv blocked due to full queue Number of times idmp_assem blocked due to full queue CPU run time taken by bmb_main CPU run time taken by pipeline threads Histogram of number of incremental snapshots present in every transfer CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

vscan_server_stat object Counter Name vscan_ops_server_la tency Size 8B Type AVERA GE Display Unit MSECS Description Average latency for virus scan operations in Priority CM_DIAG Depe vscan_ops_server_latenc

Performance Management Design Guide

129 of 138

3/7/2008

milliseconds Total time spent waiting for virus scan requests to be used as a base counter for vscan average latency calculation Count of scan request RPCs issued to the vscan server

vscan_ops_server_la tency_base

8B

DELTA

NONE

CM_DIAG

scanrequests_total_s erver scanrequests_to_ser ver_rate scancompletions_fro m_server_rate

8B

RAW

NONE

CM_DIAG

8B

RATE

Rate of scan requests PER_SE issued to the vscan C server Rate of scan PER_SE completions received C from the vscan server NONE Count of scan requests which did not successfully conclude Count of scan completions which reported viruses Count of status RPCs issued for requests which timed out Count of requests which timed out Most simultaneous scan requests to this vscan server Count of scan requests in progress to this vscan server

CM_DIAG

8B

RATE

CM_DIAG

scanfailures_total_se 8B rver virus_detections_tot al_server scanrequest_timeout _inquiries_server scanrequest_timeout _abort_server scanrequests_max_s erver 8B

RAW

CM_DIAG

RAW

NONE

CM_DIAG

8B 8B 8B

RAW RAW RAW

NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

scanrequests_current 8B _server scanrequests_oldest _server scantime_total_serv er scantime_count_ser ver 8B

RAW

NONE

CM_DIAG

RAW

Age of oldest request MSECS in progress to this vscan server Total time spent for MSECS virus scans in milliseconds NONE Count of virus scans for scan_time_total

CM_DIAG

8B 8B

RAW RAW

CM_DIAG CM_DIAG

Performance Management Design Guide

130 of 138

3/7/2008

vscan_server_latenc y vscan_server_latenc y_base

8B

AVERA GE DELTA

Average latency for MSECS virus scans in milliseconds NONE Base counter for scantime latency calculation

CM_DIAG

vscan_server_latency_ba

8B

CM_DIAG

vscan_stats object Counter Name scanrequests_total scanrequests_started scanrequests_completed Size 8B 8B 8B Type RAW RATE RATE Display Unit NONE PER_SEC PER_SEC Description Count of scan requests issued Rate of scan requests issued by the filer Rate of scan completions received from the vscan server Count of scan requests which did not successfully conclude Count of scan completions which reported viruses Count of client accesses which might cause a virus scan Count of status RPCs issued for requests which timed out Count of requests with at least one status RPCs issued for timeout Count of requests which timed out Count of scans avoided because file is marked already scanned Priority CM_DIAG CM_DIAG CM_DIAG Depends on

scanfailures_total

8B

RAW

NONE

CM_DIAG

virus_detections_total

8B

RAW

NONE

CM_DIAG

scanrequests_needed_total

8B

RAW

NONE

CM_DIAG

scanrequest_timeout_inquiries request_timeout_inquiries_uniq ue scanrequest_timeout_abort scanrequests_already

8B

RAW

NONE

CM_DIAG

8B 8B 8B

RAW RAW RAW

NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

scanrequests_already_reset

8B

RAW

NONE

Count of files whose already scanned status was CM_DIAG cleared 131 of 138 3/7/2008

Performance Management Design Guide

scanrequests_duplicate

8B

RAW

NONE

Count of files accessed while a scan was already in progress No scan for file access that would normally cause a scan Client request denied because no scan could be performed Most simultaneous scan requests delayed because all vscan servers are busy Total scan requests delayed because all vscan servers are busy Total scan requests returned to the delay list a second time because all vscan servers are busy Count of disconnections initiated by vscan server Count of disconnections initiated by filer Total time spent for virus scans in milliseconds Count of virus scans for scan_time_total Count of scan requests in progress Age of oldest request in progress Active scan requests delayed because all vscan servers are busy Age of oldest active scan request delayed because all vscan servers are busy Average latency for virus scans in milliseconds Base counter for vscan

CM_DIAG

scanrequests_noscan

8B

RAW

NONE

CM_DIAG

scanrequests_noscan_deny

8B

RAW

NONE

CM_DIAG

scanrequests_throttled_max

8B

RAW

NONE

CM_DIAG

scanrequests_throttled_total

8B

RAW

NONE

CM_DIAG

scanrequests_throttled_again

8B

RAW

NONE

CM_DIAG

disconnect_by_vscanserver disconnect_by_filer scantime_total scantime_count scanrequests_current scanrequests_oldest scanrequests_throttled_current

8B 8B 8B 8B 8B 8B 8B

RAW RAW RAW RAW RAW RAW RAW

NONE NONE MSECS NONE NONE MSECS NONE

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

scanrequests_throttled_oldest scantime_avg_latency scantime_latency_base

8B 8B 8B

RAW AVER AGE DELT

MSECS MSECS NONE

CM_DIAG CM_DIAG CM_DIAG scantime_lat ency_base

Performance Management Design Guide

132 of 138

3/7/2008

scantime latency calculation

wafl object Counter Name name_cache_hit name_cache_miss name_cache_miss_long name_cache_miss_snapdir find_dir_hit find_dir_miss buf_hash_hit buf_hash_miss inode_cache_hit inode_cache_miss inode_cache_hit_nodes inode_cache_miss_nodes inode_cache_hit_tohead inode_eject_time inode_list_cnt buf_load_cnt buf_miss_cnt buf_eject_time wafl_bufs_total wafl_bufs_dirty wafl_bufs_recycle wafl_bufs_free_bdata Siz e Type Display Unit PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC PER_SEC MSECS NONE PER_SEC PER_SEC MSECS NONE NONE NONE NONE Description Name cache hits per second Name cache misses per second Name cache misses per second due to long path names Name cache misses per second due to snapdir Directory find hit per second Directory find misses per second Buffer hash hits per second Buffer hash misses per second Inode cache hit count per second Inode cache miss count per second Inode cache hit nodes count per second Inode cache miss nodes count per second CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Prior

8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RATE 8B RAW 4B RAW 8B RATE 8B RATE 8B RAW 4B RAW 4B RAW 4B RAW 4B RAW

Inode cache hit moved to head count CM_DIAG per second Instantaneous residence time of the last inode ejected in milliseconds inode list counts Buffer cache hit count per second Buffer cache miss count per second Instantaneous residence time of the last buffer ejected in milliseconds wafl_bufs_total wafl_bufs_dirty wafl_bufs_recycle wafl_bufs_free_bdata CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

133 of 138

3/7/2008

wafl_bufs_empty wafl_bufs_available wafl_bufs_available_for_c p wafl_vbufs_total wafl_vbufs_dirty wafl_vbufs_recycle wafl_vbufs_free wafl_vbufs_unuseable wafl_vbufs_available bufs_kmem bufs_raid bufs_stolen_by_backup cp_count total_cp_msecs cp_phase_times write_alloc_nwips write_alloc_nbufs write_alloc_nbufs_other wafl_total_blk_reads wafl_blk_reads wafl_total_blk_readaheads wafl_blk_readaheads wafl_total_blk_writes wafl_blk_writes hda_avoidance_interval hda_hot_rg hda_deg_rg

4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 4B RAW 8B DELTA 8B DELTA 8B PERCT 8B DELTA 8B DELTA 8B DELTA 8B RATE 8B PERCT 8B RATE 8B PERCT 8B RATE 8B PERCT 8B RAW 8B RAW 8B RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE MSECS PERCT NONE NONE NONE PER_SEC PERCT PER_SEC PERCT PER_SEC PERCT NONE NONE NONE

wafl_bufs_empty wafl_bufs_available wafl_bufs_available_for_cp wafl_vbufs_total wafl_vbufs_dirty wafl_vbufs_recycle wafl_vbufs_free wafl_vbufs_unuseable wafl_vbufs_available bufs_kmem bufs_raid bufs stolen by backup Array of counts of different types of CPs Milliseconds spent in CP Array of percentage time spent in different phases of CP Total number of wips cleaned Total number of bufs cleaned during xphase Total number of bufs cleaned outside xphase Total blocks read per second Array of percentage of wafl blocks read by type Total blocks readahead per second Array of percentage of wafl blocks readahead by type Total blocks written per second Array of percentage of wafl blocks written by type Hotdisk avoidance interval Number of hot RGs at the start CP that will be skipped Number of degraded RGs in CP that

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

134 of 138

3/7/2008

will be skipped hda_hot_skip_count hda_deg_skip_count markers_dyn_alloc markers_freed markers_mr_alloc wrsv_rgs_full_low wrsv_rgs_full_high wrsv_rgs_full wrsv_vol_full_low wrsv_vol_full_high wrsv_vol_resets avg_wafl_msg_latency wafl_msg_total avg_non_wafl_msg_latenc y non_wafl_msg_total suspend_because new_msg_cnt restart_msg_cnt complete_in_waffi_cnt 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B RAW 8B AVERAG E NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE MSECS PER_SEC MSECS PER_SEC NON_ZER O NONE NONE NON_ZER O Number of times HDA code ran for hot disk Number of times HDA code ran for degraded disk Number of markers dynamically allocated Number of markers freed Number of markers dyn alloc'ed in the walloc/requeue path CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Number of times a raid group has hit CM_DIAG full_low for walloc Number of times a raid group has hit CM_DIAG full_high for walloc Number of times a raid group has been full for walloc Number of times a volume has hit full_low Number of times a volume has hit full_high Number of times a volume has reset its walloc thresholds Average turnaround time for wafl messages in milliseconds Total number of wafl messages per second Average turnaround time for nonwafl messages in milliseconds Total number of non-wafl messages per second Suspend Reason counts per message type New Message Counts per message type Message Restart Counts per message type Completions in Waffinity per message type CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

8B RATE 8B AVERAG E

8B RATE 4B DELTA 8B DELTA 8B DELTA 8B DELTA

CM_DIAG|CM_ASU

CM_DIAG|CM_ASU

CM_DIAG|CM_ASU

Performance Management Design Guide

135 of 138

3/7/2008

extent-alloc-fail extent-async-load extent-async-load-readreallocate extent-max-messages-load extent-max-savedmessages extent-cp-old extent-message-limit extent-max-readers extent-reader-limit extent-no-readaheadsuspend extent-no-readahead-fail extent-no-space-fail extent-modify extent-modify-keep-vvbn pvbn-alias keep_vvbn_requested keep_vvbn_done keep_vvbn_to_real quota_stats quota_pool_buf_count quota_calls

8B DELTA 8B DELTA 8B DELTA 8B RAW 8B RAW 8B DELTA 8B DELTA 8B RAW 8B DELTA 8B DELTA 8B DELTA 8B DELTA 8B DELTA 8B DELTA 8B DELTA 8B DELTA 8B DELTA 8B DELTA 4B RAW 4B RAW 4B RAW

NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE NONE

Number of faillures allocating extent messages Number of async extent loads Number of async extent read reallocate loads Maximum number of outstanding extent messages Maximum number of saved extent messages Number of extent loads associated with an old CP Number of extent reads skipped due to message limit Number of extent messages suspended due to reader limit Number of extent messages suspended due to no readahead Number of extent messages failed due to no readahead Number of extent messages failed due to no disk space Number of normal block modifies Number of keep_vvbn block modifies Number of cache blocks with significant duplicate PVBN Number of keep-vvbn block writes requested Number of keep-vvbn block writes completed Number of keep-vvbn block writes converted to real writes Global Quota stats Global Quota pool buffer count Number of Quota

CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

Number of extent concurrent readers CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG

CM_PRIV_DIAG|CM M_TEST

CM_PRIV_DIAG|CM M_TEST

CM_PRIV_DIAG|CM

Performance Management Design Guide

136 of 138

3/7/2008

(SNMP/ZAPI/other) calls

M_TEST

cifs_watch object Counter Name watch_cache_hits Size 8B Type RAW Display Unit NONE Description Number of times an inode watch cache was valid Number of times an inode watch cache was updated Number of times an inode watch cache was created Priority CM_DIAG Depends on

watch_cache_updates

8B

RAW

NONE

CM_DIAG

watch_cache_creates watch_ancestor_root_wal ks

8B

RAW

NONE

CM_DIAG

8B

RAW

NONE

Number of times watch ancestor walks to CM_DIAG the root were performed Number of times single level watch ancestor walks were performed CM_DIAG

watch_ancestor_onelevel_ 8B walks watch_ancestor_path_wal ks watch_ancestor_path_cac he_hits watch_ancestor_path_len gth_average watch_post_cache_hits

RAW

NONE

8B

RAW

NONE

Number of ancestor walks to compute watch CM_DIAG paths Number of times cached watch paths were good Average path lengths of watch paths Number of times watch post processing was averted because of cache hits Number of walks to compute watch posts Number of watch posts Number of directories CM_DIAG CM_DIAG

8B 8B

RAW RAW

NONE NONE

8B

RAW

NONE

CM_DIAG

watch_post_walks watch_posts watch_dirs

8B 8B 4B

RAW RAW RAW

NONE NONE NONE

CM_DIAG CM_DIAG CM_DIAG

Performance Management Design Guide

137 of 138

3/7/2008

currently watched

zodiac object Counter Name cycles utilization read_ops read_chain read_thruput read_latency read_errors write_ops write_chain write_thrupu t write_latenc y write_errors received completed Size 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B 8B Type DELTA NODISP PERCT RATE AVERAGE RATE AVERAGE RAW RATE AVERAGE RATE AVERAGE RAW RAW RAW Display Unit Description Priority CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG CM_DIAG write_ops write_ops read_ops read_ops cycles Depends on

DELTA Cycle counter for time NODISP base PERCT PER_SE C NONE Percentage utilization of the I/O channel Number of read operations completed per second Average descriptors per read operation

MB_PER Throughput for read _SEC operations per second USECS NONE PER_SE C NONE Average read latency per operation in microseconds Total number of read errors Number of write operations completed per second Average descriptors per write operation

MB_PER Thruoughput for write _SEC operations per second USECS NONE NONE NONE Average write latency per operation in microseconds Total number of write errors Total number of operations received Total number of operations completed

Performance Management Design Guide

138 of 138

3/7/2008

Das könnte Ihnen auch gefallen