はじめに
AIX 7.3 TL4 より、これまで手動でダウンロードが必要だったパフォーマンス診断ツール perfpmr が、AIX の標準機能として正式に組み込まれました。
地味なアップデートに見えますが、AIX 管理者・自動化スクリプトの観点では大きな変化です。
本記事ではその背景・変更点・実際の使い方をまとめます。
動画
当記事を元にした動画を作成しました。理解の一助にお役立てください。
perfpmr とは
perfpmr（Performance PMR） は、AIX システムのパフォーマンス問題を診断・トラブルシューティングするためのユーティリティです。
CPU 枯渇・メモリページング・ディスク I/O 遅延・ネットワーク性能など、多角的なシステム統計を一括収集し、IBM サポートへのエスカレーション時に必要なデータを効率よく取得できます。
主な収集データ：
| カテゴリ | コマンド・ツール |
|---|---|
| システム統計 |
vmstat, iostat, netstat, sar, mpstat
|
| トレース |
trace, lock trace, filemon, tprof
|
| 構成情報 |
lscfg, lslpp, snap など |
| ネットワーク |
iptrace, tcpdump
|
README を読む
ツールは /opt/IBM/perfpmr/ にインストールされており、同梱の README に詳細な手順が記載されています。
# cat /opt/IBM/perfpmr/README
README の要点を以下にまとめます。
| 項目 | 内容 |
|---|---|
| 対応環境 | AIX 7.3 TL04 以降 / VIOS 4.1.2.0 以降 |
| インストールパス | /opt/IBM/perfpmr/ |
| 実行権限 | root 必須 |
| 推奨空き容量 | 45MB × 論理CPU数 以上 |
| FS の制約 | リモートマウント FS での実行禁止（iptrace ハングの恐れ） |
| データ提出先 | IBM eCuRep（FTP / HTTPS） |
主な注意事項：
- データ収集自体がわずかなオーバーヘッドを発生させる
- トレース出力は 数百 MB 規模 になる場合がある
- HACMP 環境では Dead Man Switch と同一ディスクを使用しないこと、または監視タイムアウトを事前に延長すること
- iptrace はパケットレートが高い環境でネットワークスループットに影響を与えるため、不要な場合は
-Tオプションで省略可能 - 収集後は
PROBLEM.INFOに問題の背景を記入して IBM に提出する
README 原文（クリックで展開）
---------------------------------------------------------
This perfpmr package contains a number of performance tools and some instructions. Some of these products available with AIX. These tools are now officially part of AIX, and we are providing support for them.
AIX PERFORMANCE DATA COLLECTION PROCESS
---------------------------------------------------------
Note: The act of collecting performance data may add a small bit of extra overhead. There may be a large volume of trace data (depending on how many logical cpus are busy) that is written to the filesystem; this may impact the HACMP Dead Man Switch monitor if it's using the same disk as where the data is being collected.If that's the case, then either use a different filesystem/disk for the data collection or extend the Dead Man Switch timeout prior to collecting perfpmr data to avoid accidental failovers.
TABLE OF CONTENTS
-----------------
I. INTRODUCTION
II. HOW TO OBTAIN TOOLS
III. HOW TO COLLECT DATA FOR AN AIX PERFORMANCE PROBLEM
IV. HOW TO SEND DATA TO IBM
I. INTRODUCTION
This package contains a set of tools and instructions for collecting the data needed to analyze a AIX performance problem. This tool set runs on AIX 7.3 TL04 and VIOS level 4.1.2.0 onwards.
II. HOW TO OBTAIN TOOLS
The perfpmr tools is available inside perfpmr directory on both AIX and VIOS.
Below is the path of perfpmr tool:
=====================================
/opt/IBM/perfpmr/
This version should run fine on AIX 7.3 TL04 and VIOS level 4.1.2.0 onwards.
login as root or use the 'su' command to obtain root authority
III. HOW TO COLLECT DATA FOR AN AIX PERFORMANCE PROBLEM
A. Purpose:
1. This section describes the set of steps that should be followed to collect performance data.
2. The goal is to collect a good base of information that can be used by AIX technical support specialists or development lab programmers to get started in analyzing and solving the performance problem. This process may need to be repeated after analysis of the initial set of data is completed.
B. Collection of the Performance Data on Your System
1. Detailed System Performance Data:
Detailed performance data is required to analyze and solve a performance problem. Follow these steps to invoke the supplied shell scripts:
NOTE: You must have root user authority when executing these shell scripts.
a. Create a data collection directory and 'cd' into this directory.
Allow at least 45MB*#of_logicalcpus of unused space in whatever file system is used.
*IMPORTANT* - DO NOT COLLECT DATA IN A REMOTELY MOUNTED FILESYSTEM SINCE IPTRACE MAY HANG
For example using /tmp filesystem:
# mkdir /tmp/perfdata
# cd /tmp/perfdata
If the disks for /tmp aren't fast, then it's preferable to use non-rootvg disks.
b. HACMP users:
If HACMP's deadman switch monitor is checking response time of an I/O to a disk,then keep in mind that perfpmr will dump out possibly hundreds of megabytes of data near the beginning of perfpmr when the trace script is run. This may impact the HACMP monitor thread if it is using the same disk as where the perfpmr data is collected. It is generally recommended to lengthen the HACMP deadman switch interval while performance data is being collected if that is the case.
c. Collect the 'standard' PERFPMR data for 600 seconds (this is the default)
Start the data collection while the problem is already occurring with the command:
/directory_where_perfpmrscripts_are_installed/perfpmr.sh
The perfpmr.sh shell provided will:
- immediately collect a 5 second trace (trace.sh 5) and then a 5-second lock trace
- there will be a 60-second trace on just one specific event ID after that
- collect 600 seconds of general system performance data (monitor.sh 600).
- collect hardware and software configuration information (config.sh).
In addition, if it finds the following programs available
in the current execution path, it will:
- collect 10 seconds of iptrace information (iptrace.sh 10)
- collect 10 seconds of filemon information (filemon.sh 10)
- collect two 60 seconds of tprof information (tprof.sh 60)
- Note that iptrace can have a noticeable impact on network throughput
when the packet rate is high. So if a 10-second period of impact to network throughput is unacceptable, then use the -T option of perfpmr.sh to omit iptrace collection.
NOTE:
Since performance problems may mask other problems, it is not uncommon to fix one issue and then collect more data to work on another issue.
d. Answer the questions in the text file called 'PROBLEM.INFO' in the data collection directory created above. This background information about your problem helps us better understand what is going wrong.
IV. HOW TO SEND THE DATA TO IBM.
A. Combine all the collected data into a single binary 'tar' file and compress it:
Put the completed PROBLEM.INFO in the same directory where the data was collected (ie. /tmp/perfdata in the following example). Change to the parent directory, and use the tar command as follows:
Package the data in one of the following ways:
1)
# cd /tmp/perfdata
# cd ..
# pax -xpax -vw perfdata | gzip -c > salesforceticket#.pax.gz
or
2) cd /tmp; perfpmr.sh -o perfdata -z salesforceticket#.pax.gz
B. Submission of testcase to IBM:
Blue Diamond clients should upload data only to the Blue Diamond servers. For all other clients, please use the instructions at this link to upload data to IBM:
https://www.ibm.com/support/pages/enhanced-customer-data-repository-ecurep-send-data-ftp
Click on the link for the upload method of your choice (ftp, https, Java)
Data placed on the server listed below cannot be accessed by unauthorized personnel.
All servers enforce authentication. For Web access, you need to authenticate with your IBMid. For FTPS and SFTP servers, you need to create an IBM Support File Transfer ID. The IBM Support File Transfer ID is valid until revoked.
https://www.ibm.com/support/pages/enhanced-customer-data-repository-ecurep-send-data-ftp
Userid: your support id
password: your password
'cd toibm/aix'
'bin'
'put SF#.BRANCH#.COUNTRY#.pax.gz'
ex. 'TS01234567.pax.gz'
or 'TS01234567.perfpmr.hostname1.pax.gz'
'quit'
If the transfer fails with an error, it's possible that a file already exists by the same name on the ftp server. In this case, add something to the name of the file to differentiate it from the file already on the ftp site (ex. TS01234567.july18.pax.gz).
Notify your IBM customer representative you have submitted the data.
AIX 7.3 TL4 での変更点
TL4 以前：毎回 FTP からダウンロードが必要
TL4 以前は、perfpmr を使用するにはダウンロードと導入が必要でした。
# OS バージョンを確認して該当スクリプトをダウンロード
ftp https://public.dhe.ibm.com/aix/tools/perftools/perfpmr/perf73/perf73.tar.Z
uncompress perf73.tar.Z
tar -xvf perf73.tar
./perfpmr.sh
自動化スクリプトに組み込む場合も、OS バージョンの判定 → ダウンロード → 展開 → 実行、という複雑なロジックが必要でした。
TL4 以降：固定パスに統合済み
bos.perf.tools フィールドセットに統合され、OS インストール時に自動でセットアップされます。
# perfpmr の実行
/opt/IBM/perfpmr/perfpmr.sh
変更点まとめ
| 項目 | TL4 以前 | TL4 以降 |
|---|---|---|
| 入手方法 | IBM FTP から手動ダウンロード | OS インストール時に自動インストール済み |
| インストール場所 | 任意ディレクトリ |
/opt/IBM/perfpmr/ に固定 |
| バージョン管理 | ユーザーが個別に管理 | AIX 更新（TL/SP）経由で自動提供 |
| 自動化スクリプト | DL → 展開 → 実行の複雑なロジックが必要 | 固定パスを直接コールするだけ |
| サポート状態 | 非公式（AS-IS 提供） | OS レベルで正式サポート |
実際の使い方
前提条件
- OS バージョン：AIX 7.3 TL4 以降
- 権限：root 必須
- 空き容量：
45MB × 論理CPU数以上 - リモートマウントされた FS での実行は禁止（iptrace がハングする可能性あり）
ステップ 1：データ収集ディレクトリの作成
# mkdir /tmp/perfdata
# cd /tmp/perfdata
ステップ 2：問題発生中に perfpmr を実行
パフォーマンス低下が発生している 最中 に実行します。デフォルトは 600 秒（10 分） のサンプリングです。
# /opt/IBM/perfpmr/perfpmr.sh
秒数を変更したい場合：
# /opt/IBM/perfpmr/perfpmr.sh 300 # 5分の場合
iptrace のネットワーク影響を避けたい場合は -T オプションを使用します。
/opt/IBM/perfpmr/perfpmr.sh -T
ステップ 3：データの圧縮と IBM への提出
# cd /tmp
# perfpmr.sh -o perfdata -z TS#チケット番号#.pax.gz
稼働確認
AIX 7.3 TL4-00-2546 環境でオプションなし実行してみます。
# oslevel -s
7300-04-00-2546
実行結果は 約 19 分 で完了しました。実行中は以下のフェーズが順番に走ります：
| フェーズ | 処理内容 | 所要時間（目安） |
|---|---|---|
| 1 |
trace.sh（5秒トレース + ロックトレース） |
約 15 秒 |
| 2 |
trace_419.sh（特定イベント 60秒トレース） |
約 60 秒 |
| 3 |
monitor.sh（600秒のシステム全体統計収集） |
約 633 秒 |
| 4 |
iptrace.sh / tcpdump.sh（各 10秒） |
約 36 秒 |
| 5 |
filemon.sh（10秒） |
約 12 秒 |
| 6 |
tprof.sh（60秒 × 2回） |
約 124 秒 |
| 7 |
config.sh（HW/SW 構成情報収集 + snap） |
約 252 秒 |
実行ログ全文
time /opt/IBM/perfpmr/perfpmr.sh の実行ログ（クリックで展開）
# time /opt/IBM/perfpmr/perfpmr.sh
PERFPMR: Memory required: 260000000 bytes (247 MB) noncomputational_mem_available=2029080576 bytes (1935) MB
(C) COPYRIGHT International Business Machines Corp., 2000,2001,2002,2003,2004-2011
21:10:06-04/24/26 : perfpmr.sh begin
PERFPMR: hostname: testaix7304
PERFPMR: systemid: IBM,06785CA21
PERFPMR: perfpmr.sh Version Universal 2025/09/24 oslevel: 7.3
PERFPMR: current directory: /tmp/perfdata
PERFPMR: perfpmr tool directory: /opt/IBM/perfpmr
PERFPMR: Parameters passed to perfpmr.sh:
PERFPMR: Data collection started in foreground (renice -n -20)
PERFPMR: Current timezone: CST6CDT
21:10:06-04/24/26 : PERFPMR: executing perfpmr_perfpmr_premonitor -p 1 -t 1 stanza
PERFPMR: Executing /opt/IBM/perfpmr/perfpmr_premonitor.sh -p 1 -t 1
Starting perfstat_trigger at Fri Apr 24 21:10:06 CDT 2026
Starting tcpstat at Fri Apr 24 21:10:06 CDT 2026
21:10:07-04/24/26 : PERFPMR: executing perfpmr_allocate_tracebuffers stanza
21:10:07-04/24/26 : PERFPMR: executing perfpmr_trace -k 10e,254,116,117 -L 20000000 -T 20000000 -I -c -E7 5 stanza
PERFPMR: Executing /opt/IBM/perfpmr/trace.sh.sh -p -k 10e,254,116,117 -L 20000000 -T 20000000 -I -c -E7 5
TRACE.SH: Stopping any previously running traces
/dev/systrctl: No such device or address
TRACE.SH: trcstop -s not supported for non-circ files. Retrying without -s option
/dev/systrctl: No such device or address
TRACE.SH: Starting trace for 5 seconds
/usr/bin/trace -p -r PURR -k 10e,254,116,117 -f -n -C all -d -L 20000000 -T 20000000 -ao trace.raw
21:10:07-04/24/26 : initializing trace buffers : execution_time: 0.153 seconds
21:10:07-04/24/26 : trcon completed
TRACE.SH: Data collection started
TRACE.SH: trace data collection is being stopped
21:10:12-04/24/26 : trcoff completed
21:10:12-04/24/26 : TRACE.SH: Suspending component trace
21:10:12-04/24/26 : TRACE.SH: component trace suspended
21:10:12-04/24/26 : trcstop : execution_time: 0.0210000000000008 seconds
TRACE.SH: Trace stopped
21:10:12-04/24/26 : trcstop completed
TRACE.SH: Enabling locktrace
lock tracing enabled for all classes
TRACE.SH: Changing lock trace level from <3> to <7> temporarily
TRACE.SH: Starting trace for 5 seconds
/usr/bin/trace -p -r PURR -j 234,101,104,106,10C,10E,112,113,134,139,465,46D,46E,5D8,606,607,608,609,1044,4AF,419,11F,107,200,102,103,102F,4B0 -f -n -C all -d -L 20000000 -T 20000000 -ao trace.raw.lock
21:10:12-04/24/26 : initializing lock trace buffers : execution_time: 1.115 seconds
21:10:13-04/24/26 : trcon completed
TRACE.SH: Data collection started
TRACE.SH: trace data collection is being stopped
21:10:18-04/24/26 : trcoff completed
21:10:18-04/24/26 : trcstop : execution_time: 0.0800000000000001 seconds
TRACE.SH: Trace stopped
21:10:18-04/24/26 : trcstop completed
TRACE.SH: Disabling locktrace
lock tracing disabled for all classes
TRACE.SH: Setting locktrclevel of <7> back to previous level of <3>
TRACE.SH: Trcnm data is in file trace.nm
TRACE.SH: /etc/trcfmt saved in file trace.fmt
TRACE.SH: Binary trace data files are in trace.raw.lock*
TRACE.SH: Trcnm data is in file trace.nm
TRACE.SH: /etc/trcfmt saved in file trace.fmt
TRACE.SH: skipping collection of gennames/gensyms/inode data until config.sh
TRACE.SH: Resuming component trace from suspended state
21:10:22-04/24/26 : trace.sh completed: execution_time: 15.262 seconds
21:10:22-04/24/26 : PERFPMR: executing perfpmr_comptrace -L 20000000 -T 20000000 -d -s 5 stanza
PERFPMR: skipping comptrace collection
21:10:22-04/24/26 : PERFPMR: executing perfpmr_trace_419 -L 10000000 -T 5000000 60 stanza
PERFPMR: Executing /opt/IBM/perfpmr/trace_419.sh -L 10000000 -T 5000000 60
21:10:22-04/24/26 : trace_419.sh
/usr/bin/trace -C all -j 419 -L 10000000 -T 5000000 -ano trc_419.raw
21:11:22-04/24/26 : trace_419 has been stopped
21:11:22-04/24/26 : trace_419.sh completed: execution_time: 60.194 seconds
21:11:22-04/24/26 : PERFPMR: executing perfpmr_monitor -I 0 -N 0 -S 0 600 stanza
PERFPMR: Executing /opt/IBM/perfpmr/monitor.sh -I 0 -N 0 -S 0 600
21:11:22-04/24/26 : monitor.sh -I
MONITOR: Capturing initial lsps, svmon, and vmstat data
21:11:23-04/24/26 : MONITOR: getting vnic stats into vnicstats.before
21:11:23-04/24/26 : MONITOR: getting pile stats into pile.before
21:11:27-04/24/26 : svmon -G : execution_time: 0.004 seconds
21:11:27-04/24/26 : svmon -P -O mapping=on,maxbufsize=32MB : execution_time: 0.024 seconds
21:11:27-04/24/26 : svmon -S -O pidlist=on,maxbufsize=32MB svmon internal error: 0x0001 : svm_segment failed due to select_vsid :occurence = 1 errno = 9
: execution_time: 0.018 seconds
21:11:27-04/24/26 : svmon.sh completed : execution_time: 0.062 seconds
21:11:27-04/24/26 : fcstat before on each fc adapter
21:11:28-04/24/26 : fcstats completed
21:11:28-04/24/26 : lsmpio before
21:11:28-04/24/26 : IPSEC before
MONITOR: Starting perf_xtra programs: initsleep=0 count=0 sleep=0
21:11:28-04/24/26 : MONITOR: Starting system monitors for 600 seconds.
Starting <mpstat> monitoring
21:11:28-04/24/26 : perfxtra.sh completed: execution_time: 0.002 seconds
Starting <ps> monitoring
Starting <nfs> monitoring
Starting <netstat> monitoring
Starting <24x7count> monitoring
Starting <kdb> monitoring
Starting <emstat> monitoring
Starting <tstat> monitoring
Starting <sar> monitoring
Starting <iostat> monitoring
Starting <aiostat> monitoring
Starting <ventstat> monitoring
Starting <fcstat2> monitoring
Starting <iomon> monitoring
Starting <foldstat> monitoring
Starting <powerstat> monitoring
Starting <lparstat> monitoring
Starting <vmstat> monitoring
iostat: No tapes found in the system.
iostat: 0551-157 Asynchronous I/O not configured on the system.
lparstat: 0551-016 Command not supported on this Platform.
Starting <pprof> monitoring
21:11:35-04/24/26 : MONITOR: Waiting for measurement period to end....
21:21:50-04/24/26 : MONITOR: Capturing final lsps, svmon, and vmstat data
21:21:50-04/24/26 : MONITOR: getting vnic stats into vnicstats.after
21:21:50-04/24/26 : MONITOR: getting pile stats into pile.after
21:21:53-04/24/26 : svmon -G : execution_time: 0.004 seconds
21:21:54-04/24/26 : svmon -P -O mapping=on,maxbufsize=32MB : execution_time: 0.023 seconds
21:21:54-04/24/26 : svmon -S -O pidlist=on,maxbufsize=32MB svmon internal error: 0x0001 : svm_segment failed due to select_vsid :occurence = 1 errno = 9
: execution_time: 0.018 seconds
21:21:54-04/24/26 : svmon.sh completed : execution_time: 0.06 seconds
21:21:54-04/24/26 : fcstat after on each fc adapter
21:21:55-04/24/26 : fcstats completed
21:21:55-04/24/26 : lsmpio after
21:21:55-04/24/26 : IPSEC after
MONITOR: Generating reports....
MONITOR: Network reports are in netstat.int and nfsstat.int
MONITOR: Monitor reports are in monitor.int and monitor.sum.
21:21:55-04/24/26 : monitor.sh completed: execution_time: 632.923 seconds
21:21:55-04/24/26 : PERFPMR: executing perfpmr_iptrace -L 0 -m500000 10 stanza
PERFPMR: Executing /opt/IBM/perfpmr/iptrace.sh -L 0 -m500000 10
Checking current packet rate for 5 seconds
Average packet rate of <12> is less than maxrate threshold of <500000>
iptrace collection will proceed
IPTRACE: Starting iptrace for 10 seconds....
0513-059 The iptrace Subsystem has been started. Subsystem PID is 10223938.
0513-044 The iptrace Subsystem was requested to stop.
IPTRACE: iptrace collected.
IPTRACE: Binary iptrace data is in file iptrace.raw : execution_time: 10.022 seconds
21:22:12-04/24/26 : PERFPMR: executing perfpmr_tcpdump -m500000 10 stanza
PERFPMR: Executing /opt/IBM/perfpmr/tcpdump.sh -m500000 10
TCPDUMP: Starting tcpdump for 10 seconds....
Checking current packet rate for 5 seconds
TCPDUMP: tcpdump collected.
TCPDUMP: Binary tcpdump data is in file tcpdump.raw.en0
TCPDUMP COLLECTION : execution_time: 10.015 seconds
21:22:28-04/24/26 : PERFPMR: executing perfpmr_filemon -T 10000000 -O detailed,all -D 10 stanza
PERFPMR: Executing /opt/IBM/perfpmr/filemon.sh -T 10000000 -O detailed,all -D 10
FILEMON: Starting filesystem monitor for 10 seconds....
21:22:28-04/24/26 : Starting trace -C all -j 139,134,107,10B,15B,163,19C,1C9,221,222,228,232,2A1,2A2,45B,5D8,AB2 -L 10000000 -T 10000000 -andfo filemon_trace.raw : execution_time: 0.14 seconds
21:22:29-04/24/26 : trcon initiated : execution_time: 0.00299999999999989 seconds
FILEMON: tracing started
21:22:39-04/24/26 : trcstop initiated : execution_time: 0.00900000000000034 seconds
21:22:39-04/24/26 : FILEMON: trcstop completed
FILEMON: Generating report....
21:22:39-04/24/26 : filemon report generating .. 21:22:39-04/24/26 : running gensyms : execution_time: 1.123 seconds
21:22:40-04/24/26 : FILEMON: filemon.sh post processing completed
21:22:40-04/24/26 : FILEMON: filemon.sh removing trace files
21:22:40-04/24/26 : FILEMON: filemon.sh completed : execution_time: 12.307 seconds
21:22:40-04/24/26 : PERFPMR: executing perfpmr_tprof -f10 60 stanza
PERFPMR: Executing /opt/IBM/perfpmr/tprof.sh -f10 60
TPROF: Starting tprof for 60 seconds....
Creating /tmp/perfdata/tprof_data/tprofdata_purrbased
21:22:41-04/24/26 : Running purr-based tprof -v -R -T 20000000 -l -r tprof -F -A all -x /usr/bin/sleep 60 : execution_time: 61.417 seconds
/tmp/perfdata/tprof_data
Creating /tmp/perfdata/tprof_data/tprofdata_tickbased
TPROF: Starting tprof with no PURR for 60 seconds....
21:23:42-04/24/26 : Running tick-based tprof -v -T 20000000 -l -r tprof_nopurr -F -A all -x /usr/bin/sleep 60 : execution_time: 61.404 seconds
/tmp/perfdata/tprof_data
TPROF: Sample data collected....
21:24:44-04/24/26 : tprof -R -l -r tprof -skej : execution_time: 0.63600000000001 seconds
/tmp/perfdata/tprof_data
21:24:44-04/24/26 : tprof -l -r tprof_nopurr -zskej : execution_time: 0.640999999999991 seconds
/tmp/perfdata/tprof_data
TPROF: tprof trc and syms files are in tprof_data
21:24:45-04/24/26 : tprof collection completed
TPROF: Tprof report is in tprof.sum and tprof_nopurr.prof
21:24:45-04/24/26 : tprof.sh completed: execution_time: 124.434 seconds
21:24:45-04/24/26 : PERFPMR: executing perfpmr_hpmcount -H 5 stanza
PERFPMR: skipping hpmcount collection
21:24:45-04/24/26 : PERFPMR: executing perfpmr_pmucount -g all -o pmucount_kuyng.txt -e pmucount.stderr -P 59 1 stanza
PERFPMR: skipping pmucount collection
21:24:45-04/24/26 : PERFPMR: executing perfpmr_probevue 5 stanza
PERFPMR: skipping probevue collection
21:24:45-04/24/26 : PERFPMR: executing perfpmr_config -u 0 stanza
PERFPMR: Executing /opt/IBM/perfpmr/config.sh -u 0
21:24:45-04/24/26 : config.sh begin
CONFIG.SH: Generating SW/HW configuration
21:24:45-04/24/26 : copying ODM files : execution_time: 0.083 seconds
21:24:46-04/24/26 : do_kdb_vmm : execution_time: 2.647 seconds
21:24:48-04/24/26 : ipcs -Smqsa : execution_time: 0.00300000000000011 seconds
21:24:48-04/24/26 : lsmcode -A : execution_time: 0.365 seconds
21:24:49-04/24/26 : lspv -u : execution_time: 0.0209999999999999 seconds
21:24:49-04/24/26 : lspv : execution_time: 0.0220000000000002 seconds
21:24:49-04/24/26 : lsvg : execution_time: 0.0339999999999998 seconds
21:24:49-04/24/26 : lsvg -l : execution_time: 0.0180000000000002 seconds
21:24:49-04/24/26 : lslv lv : execution_time: 0.451 seconds
21:24:49-04/24/26 : lsattr -E -l dev : execution_time: 0.179 seconds
21:24:49-04/24/26 : getting disk queue_depth info : execution_time: 0.675999999999999 seconds
21:24:50-04/24/26 : disk queue_depth info completed
21:24:50-04/24/26 : df : execution_time: 0.00600000000000023 seconds
21:24:51-04/24/26 : /usr/bin/netstat -in -rn -D -an -c -Cn : execution_time: 0.0190000000000001 seconds
RPC: 1832-011 Program unavailable: nfso: 1831-730 cannot contact local statd.
RPC: 1832-011 Program unavailable: nfso: 1831-730 cannot contact local statd.
RPC: 1832-011 Program unavailable: nfso: 1831-730 cannot contact local statd.
RPC: 1832-011 Program unavailable: nfso: 1831-730 cannot contact local statd.
RPC: 1832-011 Program unavailable: nfso: 1831-730 cannot contact local statd.
RPC: 1832-011 Program unavailable: nfso: 1831-730 cannot contact local statd.
21:24:56-04/24/26 : getmempool.sh : execution_time: 1.822 seconds
21:24:58-04/24/26 : getvmpool.sh : execution_time: 0.678000000000001 seconds
21:24:58-04/24/26 : getj2mem.sh : execution_time: 0.289999999999999 seconds
21:24:59-04/24/26 : genkld : execution_time: 0.00400000000000134 seconds
21:24:59-04/24/26 : genkex : execution_time: 0.00299999999999834 seconds
21:24:59-04/24/26 : getevars : execution_time: 0.00499999999999901 seconds
21:24:59-04/24/26 : ls /proc/pid/cwd : execution_time: 0.00600000000000023 seconds
21:24:59-04/24/26 : errpt : execution_time: 0.0229999999999997 seconds
21:24:59-04/24/26 : emgr -lv3 > emgr.out There is no efix data on this system.
: execution_time: 0.277000000000001 seconds
21:24:59-04/24/26 : lslpp -ch : execution_time: 0.324999999999999 seconds
21:24:59-04/24/26 : instfix -ic : execution_time: 1.215 seconds
21:25:01-04/24/26 : lscfg -vp : execution_time: 0.327 seconds
lparstat: 0551-016 Command not supported on this Platform.
21:25:01-04/24/26 : lvol kdb cmd : execution_time: 0.459 seconds
21:25:01-04/24/26 : do_kdb_devsw : execution_time: 2.484 seconds
21:25:04-04/24/26 : echo vnode|kdb : execution_time: 3.642 seconds
21:25:08-04/24/26 : echo vfs|kdb : execution_time: 0.280999999999999 seconds
21:25:08-04/24/26 : collect dscrctl and NX settings : execution_time: 0.373999999999999 seconds
21:25:08-04/24/26 : echo dmpdt_chrp -i : execution_time: 0.297999999999998 seconds
21:25:08-04/24/26 : sysdumpdev -l, -e : execution_time: 0.00600000000000023 seconds
21:25:09-04/24/26 : gennames : execution_time: 0.715 seconds
cp: /etc/resolv.conf: No such file or directory
21:25:09-04/24/26 : VIO client cfg data : execution_time: 0.762 seconds
/var/perf/daily//*.nmon not found
./*.nmon not found
21:25:10-04/24/26 : getting alog information
21:25:10-04/24/26 : powermt data being collected
21:25:10-04/24/26 : db2 tids being collected : execution_time: 0.0159999999999982 seconds
21:25:11-04/24/26 : proctree -Tt : execution_time: 0.00900000000000034 seconds
21:25:11-04/24/26 : memdetails.sh -u 0 : execution_time: 17.495 seconds
21:25:28-04/24/26 : memdetails.sh completed
21:25:28-04/24/26 : application symbol tables being generated : execution_time: 0.0559999999999974 seconds
21:25:28-04/24/26 : collect logfiles : execution_time: 0.0399999999999991 seconds
21:25:28-04/24/26 : ipsec : execution_time: 0.0150000000000006 seconds
21:25:28-04/24/26 : smtctl : execution_time: 0.00300000000000011 seconds
21:25:28-04/24/26 : lruobj : execution_time: 1.087 seconds
/tmp/perfdata
21:25:29-04/24/26 : AHAFS: AHAFS is being used .. collecting ahafs config
: execution_time: 0.00800000000000267 seconds
21:25:30-04/24/26 : th |kdb : execution_time: 0.702999999999996 seconds
21:25:31-04/24/26 : running lsconf : execution_time: 1.571 seconds
21:25:32-04/24/26 : do_kdb_scb : execution_time: 5.268 seconds
21:25:38-04/24/26 : vnicstat.sh : execution_time: 0.0619999999999976 seconds
21:25:38-04/24/26 : radinfo : execution_time: 0.0110000000000028 seconds
21:25:38-04/24/26 : nvme_collect : execution_time: 2.20399999999999 seconds
21:25:40-04/24/26 : pmctl/pmcycles : execution_time: 0.00600000000000023 seconds
21:25:40-04/24/26 : vxfs : execution_time: 0.00800000000000267 seconds
21:25:40-04/24/26 : lcpu2bid : snap execution_time: 0.00300000000000011 seconds
21:25:40-04/24/26 : ipf : execution_time: 0 seconds
21:25:40-04/24/26 : efsenable : execution_time: 0.155999999999999 seconds
21:25:40-04/24/26 : snap -gGfFLktRSn -d /tmp/perfdata/snap_data
（snap の出力は長いため後述）
: snap execution_time: 193.193 seconds
21:28:53-04/24/26 : locale
21:28:53-04/24/26 : ipclimits
21:28:53-04/24/26 : lsuser -a capabilities ALL
21:28:53-04/24/26 : pmlist
Output in pmlist.allgroups.txt
21:28:54-04/24/26 : pmlist completed : execution_time: 0.04 seconds
21:28:54-04/24/26 : collecting inode data from kdb : execution_time: 3.26000000000002 seconds
21:28:57-04/24/26 : running gensyms : execution_time: 0.671999999999997 seconds
21:28:58-04/24/26 : gensyms completed
21:28:58-04/24/26 : proc/version
CONFIG.SH: Report is in file config.sum
21:28:58-04/24/26 : config.sh completed : execution_time: 252.431 seconds
PERFPMR: Data collection complete.
Stopping perfpmr_premonitor.sh
root 13435328 1 0 21:10:06 pts/0 0:00 /bin/ksh93 /opt/IBM/perfpmr/perfpmr_premonitor.sh -p 1 -t 1
In perfpmr_premonitor cleanup
PERFPMR: Data files can be archived and gzipped using:
perfpmr.sh -z filename [-o "dirs"]
filename is the name of the archive file.
An example of a typical archive filename:
/tmp/salesforceticket#.perfpmr.pax.gz
Or /tmp/salesforceticket#.host.perfpmr.pax.gz
"dirs" dirs is a list of directories or perfdata directory name enclosed in quotes.
When -o is specified, you must be in the parent directory of the data director[y\ies] specified
ex. perfpmr.sh -z /tmp/TS0000000.perfpmr.pax.gz -o perfdata
If -o not specified, all files in current directory are archived.
For ftp upload instructions, visit:
https://www.ibm.com/support/pages/enhanced-customer-data-repository-ecurep-send-data-ftp
NOTE: Blue Diamond clients should upload to the BD servers instead.
21:28:58-04/24/26 : perfpmr.sh completed : execution_time: 1134.113 seconds
real 18m54.20s
user 0m23.89s
sys 0m17.22s
生成されたファイル一覧
完了後、収集ディレクトリには大量のファイルが生成されます。
ls -l の結果（クリックで展開）
# ls -l
total 265592
-rw-r--r-- 1 root system 0 Apr 24 21:11 24x7count.csv
-rw-r--r-- 1 root system 94 Apr 24 21:11 24x7count.err
-rw-r--r-- 1 root system 202 Apr 24 21:11 24x7count.int
-rw-r--r-- 1 root system 113 Apr 24 21:11 24x7count_sh.outerr
-rw-r--r-- 1 root system 3636 Apr 24 21:21 acf.out
-rw-r--r-- 1 root system 943 Apr 24 21:25 ahafs.conf
-rw-r--r-- 1 root system 53 Apr 24 21:11 aiostat.int
-rw-r--r-- 1 root system 57051 Apr 24 21:25 alog.boot
-rw-r--r-- 1 root system 9676 Apr 24 21:25 alog.console
drwxr-xr-x 2 root system 256 Apr 24 21:25 aso_logs
drwxr-xr-x 2 root system 12288 Apr 24 21:10 comptrace_dir
-rw-r--r-- 1 root system 360435 Apr 24 21:28 config.sum
-rw------- 1 root system 24075 Apr 24 21:25 conslog.alog
-rw-r--r-- 1 root system 1214 Apr 24 21:25 crontab_l
drwxr-xr-x 2 root system 256 Apr 24 21:25 crontabs
-rw-r--r-- 1 root system 6092 Apr 24 21:10 ctctrl.err
-rw-r--r-- 1 root system 22148 Apr 24 21:25 devsw.out
-rw-r--r-- 1 root system 91189 Apr 24 21:25 devtree.out
-rw-r--r-- 1 root system 13991 Apr 24 21:24 disk_kdb.out
-rw-r--r-- 1 root system 72 Apr 24 21:24 disk_qd_list
-rw-r--r-- 1 root system 372 Apr 24 21:24 disk_qdepth.out
-rw-r--r-- 1 root system 2492 Apr 24 21:25 dlpi.conf
-rw-r--r-- 1 root system 263 Apr 24 21:25 efs.out
-rw-r--r-- 1 root system 208 Apr 24 21:21 eidmon.after
-rw-r--r-- 1 root system 210 Apr 24 21:11 eidmon.before
-rw-r--r-- 1 root system 91 Apr 24 21:25 emc_powermt.txt
-rw-r--r-- 1 root system 31 Apr 24 21:24 emgr.out
-rw-r--r-- 1 root system 8192 Apr 24 21:24 errlog
-rw-r--r-- 1 root system 11184 Apr 24 21:24 errpt_a
-rw-r--r-- 1 root system 503089 Apr 24 21:24 errtmplt
-rw-r--r-- 1 root system 2074 Apr 24 21:25 etc_environment
-rw-r--r-- 1 root system 2320 Apr 24 21:25 etc_filesystems
-rw-r--r-- 1 root system 3359 Apr 24 21:25 etc_inittab
-rw-r--r-- 1 root system 4541 Apr 24 21:25 etc_netsvc.conf
-r-xr-xr-x 1 root system 1814 Apr 24 21:25 etc_profile
-r-xr-xr-- 1 root system 3683 Apr 24 21:25 etc_rc
-rw------- 1 root system 14348 Apr 24 21:25 etc_sec_ldap.cfg
-rw-r----- 1 root system 1384 Apr 24 21:25 etc_security_limits
-rw-r--r-- 1 root system 344925 Apr 24 21:25 etc_services
-rw-r--r-- 1 root system 1569 Apr 24 21:21 fc.after
-rw-r--r-- 1 root system 1570 Apr 24 21:11 fc.before
-rw-r--r-- 1 root system 7567 Apr 24 21:21 fcstat.after
-rw-r--r-- 1 root system 7567 Apr 24 21:11 fcstat.before
-rw-r--r-- 1 root system 26322 Apr 24 21:21 fcstat2.csv
-rw-r--r-- 1 root system 0 Apr 24 21:11 fcstat2.err
-rw-r--r-- 1 root system 39628 Apr 24 21:21 fcstat2.out
-rw-r--r-- 1 root system 882 Apr 24 21:21 fcstat2.totals.after
-rw-r--r-- 1 root system 882 Apr 24 21:11 fcstat2.totals.before
-rw-r--r-- 1 root system 23046 Apr 24 21:22 filemon.sum
-rw-r--r-- 1 root system 103673 Apr 24 21:21 foldstat.out
-rw-r--r-- 1 root system 3033 Apr 24 21:24 genkex.out
-rw-r--r-- 1 root system 15169 Apr 24 21:24 genkld.out
-rw-r--r-- 1 root system 49876291 Apr 24 21:25 gennames.out
-rw-r--r-- 1 root system 142294 Apr 24 21:24 getevars.out
-rw-r--r-- 1 root system 929298 Apr 24 21:25 instfix.out
-rw-r--r-- 1 root system 175475 Apr 24 21:21 iomon.out
-rw-r--r-- 1 root system 181356 Apr 24 21:21 iostat.Dl
-rw-r--r-- 1 root system 76832 Apr 24 21:21 iostat.int
-rw-r--r-- 1 root system 0 Apr 24 21:11 iostat.p
-rw-r--r-- 1 root system 781 Apr 24 21:11 iostat.path
-rw-r--r-- 1 root system 205 Apr 24 21:21 iostat.sum
-rw-r--r-- 1 root system 3134 Apr 24 21:21 iostat.total
-rw-r--r-- 1 root system 672 Apr 24 21:21 ipsec.after
-rw-r--r-- 1 root system 673 Apr 24 21:11 ipsec.before
-rw-r--r-- 1 root system 2592 Apr 24 21:25 ipsec.out
-rw-r--r-- 1 root system 24373 Apr 24 21:22 iptrace.raw
-rw-r--r-- 1 root system 830 Apr 24 21:10 kdb.err
-rw-r--r-- 1 root system 54779 Apr 24 21:24 kdb.stats.out
-rw-r--r-- 1 root system 48 Apr 24 21:25 lcpu2bid.out
-rw-r--r-- 1 root system 286780 Apr 24 21:25 localhost_251121.topas
-rw-r--r-- 1 root system 5319 Apr 24 21:21 lparstat.E
-rw-r--r-- 1 root system 5979 Apr 24 21:21 lparstat.d
-rw-r--r-- 1 root system 516240 Apr 24 21:21 lparstat.int
-rw-r--r-- 1 root system 5632 Apr 24 21:21 lparstat.l
-rw-r--r-- 1 root system 1346 Apr 24 21:10 lparstat.nsp
-rw-r--r-- 1 root system 2527 Apr 24 21:11 lparstat.sum
drwxr-xr-x 2 root system 256 Apr 24 21:25 lrulist_dir
-rw-r--r-- 1 root system 3215 Apr 24 21:25 lsconf.out
-rw-r--r-- 1 root system 88460 Apr 24 21:24 lslpp.Lc
-rw-r--r-- 1 root system 10396 Apr 24 21:21 lsmpio.after
-rw-r--r-- 1 root system 10398 Apr 24 21:11 lsmpio.before
-rw-r--r-- 1 root system 424 Apr 24 21:24 lsmpio.disks.out
-rw-r--r-- 1 root system 350 Apr 24 21:21 lsps.after
-rw-r--r-- 1 root system 350 Apr 24 21:11 lsps.before
-rw-r--r-- 1 root system 187 Apr 24 21:24 lspv.uuid.out
-rw-r--r-- 1 root system 1563 Apr 24 21:25 lsrset.out
-rw-r--r-- 1 root system 66 Apr 24 21:25 lssrad.out
-rw-r--r-- 1 root system 109 Apr 24 21:28 lsuser.out
-rw-r--r-- 1 root system 0 Apr 24 21:25 lswpar.out
-rw-r--r-- 1 root system 204800 Apr 24 21:25 lvmt.log
drwxr-xr-x 2 root system 4096 Apr 24 21:25 mem_details_dir
-rw-r--r-- 1 root system 15666 Apr 24 21:24 mempools.out
-rw-r--r-- 1 root system 1791 Apr 24 21:10 memtrace.err
-rw-r--r-- 1 root system 57 Apr 24 21:24 microcode_levels.out
-rw-r--r-- 1 root system 109 Apr 24 21:25 mmfs.cfg
-rw-r--r-- 1 root system 245634 Apr 24 21:21 monitor.int
-rw-r--r-- 1 root system 51286 Apr 24 21:21 monitor.sum
-rw-r--r-- 1 root system 162819 Apr 24 21:21 mpstat.int
-rw-r--r-- 1 root system 24302 Apr 24 21:21 mpstat_h.int
-rw-r--r-- 1 root system 107100 Apr 24 21:21 mrq.out
-rw-r--r-- 1 root system 883 Apr 24 21:25 netbackup.cfg
-rw-r--r-- 1 root system 831 Apr 24 21:22 netstat.i.tmp
-rw-r--r-- 1 root system 171461 Apr 24 21:21 netstat.int
-rw-r--r-- 1 root system 253 Apr 24 21:21 netstat_sh.outerr
-rw-r--r-- 1 root system 10046 Apr 24 21:21 nfsstat.int
-rw-r--r-- 1 root system 220 Apr 24 21:21 nfsstat_sh.outerr
-rw-r--r-- 1 root system 86 Apr 24 21:25 nvme.log
-rw-r--r-- 1 root system 6633 Apr 24 21:25 nvme_data.out
drwxr-xr-x 2 root system 4096 Apr 24 21:24 objrepos
-rw-r--r-- 1 root system 7140 Apr 24 21:21 pdtstat.out
-rw------- 1 root system 5076 Apr 24 21:28 perfpmr.cfg
-rw-r--r-- 1 root system 24239 Apr 24 21:28 perfpmr.int
-rw-r--r-- 1 root system 526123 Apr 24 21:28 perfstat_trigger.premonitor.out
-rw-r--r-- 1 root system 3168 Apr 24 21:25 persistent.db
-rw-r--r-- 1 root system 5802 Apr 24 21:21 pile.after
-rw-r--r-- 1 root system 5802 Apr 24 21:11 pile.before
-rw-r--r-- 1 root system 2494 Apr 24 21:24 pile.out
-rw-r--r-- 1 root system 11470 Apr 24 21:21 pmapstat.out
-rw-r--r-- 1 root system 675740 Apr 24 21:28 pmlist.allgroups.txt
-rw-r--r-- 1 root system 3477 Apr 24 21:21 powerstat.out
-rw-r--r-- 1 root system 35220 Apr 24 21:21 ppda.out
-rw-rw---- 1 root system 87232 Apr 24 21:21 pprof.trace.raw
-rw-rw---- 1 root system 4097752 Apr 24 21:16 pprof.trace.raw-0
-rw-rw---- 1 root system 1014072 Apr 24 21:21 pprof.trace.raw-1
-rw-rw---- 1 root system 707168 Apr 24 21:21 pprof.trace.raw-2
-rw-rw---- 1 root system 590416 Apr 24 21:21 pprof.trace.raw-3
-rw-rw---- 1 root system 416872 Apr 24 21:21 pprof.trace.raw-4
-rw-rw---- 1 root system 412328 Apr 24 21:21 pprof.trace.raw-5
-rw-rw---- 1 root system 397352 Apr 24 21:21 pprof.trace.raw-6
-rw-rw---- 1 root system 395240 Apr 24 21:21 pprof.trace.raw-7
-rw-r--r-- 1 root system 4183 Apr 24 21:24 proc.cwd.out
-rw-r--r-- 1 root system 0 Apr 24 21:11 proc_sys_adapter.err
-rw-r--r-- 1 root system 32510 Apr 24 21:21 proc_sys_adapter.int.after
-rw-r--r-- 1 root system 32512 Apr 24 21:11 proc_sys_adapter.int.before
-rw-r--r-- 1 root system 15461 Apr 24 21:25 proctree.out
-rw-r--r-- 1 root system 23643 Apr 24 21:21 ps.int
-rw-r--r-- 1 root system 10865 Apr 24 21:21 ps.sum
-rw-r--r-- 1 root system 85088 Apr 24 21:21 psAfmo.after
-rw-r--r-- 1 root system 89788 Apr 24 21:11 psAfmo.before
drwxr-xr-x 2 root system 4096 Apr 24 21:21 ps_output
-rw-r--r-- 1 root system 18965 Apr 24 21:21 psa.elfk
-rw-r--r-- 1 root system 21576 Apr 24 21:11 psb.elfk
-rw-r--r-- 1 root system 8338 Apr 24 21:10 psef.begin
-rw-r--r-- 1 root system 8644 Apr 24 21:28 psef.end
-rw-r--r-- 1 root system 52954 Apr 24 21:21 psemo.after
-rw-r--r-- 1 root system 56254 Apr 24 21:11 psemo.before
-rw-r--r-- 1 root system 1063 Apr 24 21:25 radinfo.out
-r-x------ 1 root system 7438 Apr 24 21:28 reorg.sh
-rw-r--r-- 1 root system 71995 Apr 24 21:21 rqi.out
-rw-r--r-- 1 root system 285846 Apr 24 21:21 sar.bin
-rw-r--r-- 1 root system 41119 Apr 24 21:21 sar.int
-rw-r--r-- 1 root system 39008 Apr 24 21:21 sar.sum
-rw-r--r-- 1 root system 1211359 Apr 24 21:25 scb.6.0
-rw-r--r-- 1 root system 705 Apr 24 21:25 smit.log
-rw-r--r-- 1 root system 520 Apr 24 21:25 smtctl.out
drwxr-xr-x 21 root system 4096 Apr 24 21:25 snap_data
-rw-r--r-- 1 root system 3287 Apr 24 21:25 strtune.out
-rw-r--r-- 1 root system 102847 Apr 24 21:21 svmon.after
-rw-r--r-- 1 root system 93 Apr 24 21:21 svmon.after.S
-rw-r--r-- 1 root system 102847 Apr 24 21:11 svmon.before
-rw-r--r-- 1 root system 93 Apr 24 21:11 svmon.before.S
drwxr-xr-x 2 root system 256 Apr 24 21:25 syslogs
-rw-r--r-- 1 root system 159 Apr 24 21:22 tcpdump.en0.err
-rw-r--r-- 1 root system 14308 Apr 24 21:22 tcpdump.raw.en0
-rw-r--r-- 1 root system 288927 Apr 24 21:28 tcpstat.out
-rw-r--r-- 1 root system 167892 Apr 24 21:25 testaix7304_260424.topas
-rw-r--r-- 1 root system 30832 Apr 24 21:25 th.kdb.out
-rw-r--r-- 1 root system 5407 Apr 24 21:24 tprof.sum
drwxr-xr-x 4 root system 256 Apr 24 21:23 tprof_data
-rw-r--r-- 1 root system 7438 Apr 24 21:24 tprof_nopurr.prof
-rw-r--r-- 1 root system 46389 Apr 24 21:10 trace.begin.psef
-rw-r--r-- 1 root system 399 Apr 24 21:28 trace.crash.inode
-rw-r--r-- 1 root system 46769 Apr 24 21:10 trace.end.psef
-rw-r--r-- 1 root system 1920235 Apr 24 21:10 trace.fmt
-rw-r--r-- 1 root system 3977446 Apr 24 21:28 trace.j2.inode
-rw-r--r-- 1 root system 66 Apr 24 21:10 trace.lssrad.out
-rw-r--r-- 1 root system 7605 Apr 24 21:10 trace.maj_min2lv
-rw-r--r-- 1 root system 965 Apr 24 21:10 trace.mount
-rw-r--r-- 1 root system 373136 Apr 24 21:10 trace.nm
-rw-rw---- 1 root system 1757232 Apr 24 21:10 trace.raw
-rw-rw---- 1 root system 2371728 Apr 24 21:10 trace.raw-0
-rw-rw---- 1 root system 534984 Apr 24 21:10 trace.raw-1
-rw-rw---- 1 root system 75040 Apr 24 21:10 trace.raw-2
-rw-rw---- 1 root system 74384 Apr 24 21:10 trace.raw-3
-rw-rw---- 1 root system 47824 Apr 24 21:10 trace.raw-4
-rw-rw---- 1 root system 57032 Apr 24 21:10 trace.raw-5
-rw-rw---- 1 root system 47464 Apr 24 21:10 trace.raw-6
-rw-rw---- 1 root system 77480 Apr 24 21:10 trace.raw-7
-rw-rw---- 1 root system 1757520 Apr 24 21:10 trace.raw.lock
-rw-rw---- 1 root system 17965024 Apr 24 21:10 trace.raw.lock-0
-rw-rw---- 1 root system 674552 Apr 24 21:10 trace.raw.lock-1
-rw-rw---- 1 root system 212376 Apr 24 21:10 trace.raw.lock-2
-rw-rw---- 1 root system 200328 Apr 24 21:10 trace.raw.lock-3
-rw-rw---- 1 root system 140920 Apr 24 21:10 trace.raw.lock-4
-rw-rw---- 1 root system 132248 Apr 24 21:10 trace.raw.lock-5
-rw-rw---- 1 root system 114296 Apr 24 21:10 trace.raw.lock-6
-rw-rw---- 1 root system 194888 Apr 24 21:10 trace.raw.lock-7
-rw-r--r-- 1 root system 26251601 Apr 24 21:28 trace.syms
-rw-r--r-- 1 root system 19260 Apr 24 21:10 trace.tunables.out
-rw-r--r-- 1 root system 91984 Apr 24 21:10 trace_lparstatH.out
-rw-rw---- 1 root system 1756952 Apr 24 21:11 trc_419.raw
-rw-rw---- 1 root system 984512 Apr 24 21:11 trc_419.raw-0
-rw-rw---- 1 root system 137200 Apr 24 21:11 trc_419.raw-1
-rw-rw---- 1 root system 74984 Apr 24 21:11 trc_419.raw-2
-rw-rw---- 1 root system 95136 Apr 24 21:11 trc_419.raw-3
-rw-rw---- 1 root system 66624 Apr 24 21:11 trc_419.raw-4
-rw-rw---- 1 root system 68472 Apr 24 21:11 trc_419.raw-5
-rw-rw---- 1 root system 65216 Apr 24 21:11 trc_419.raw-6
-rw-rw---- 1 root system 102176 Apr 24 21:11 trc_419.raw-7
-rw-r--r-- 1 root system 49858 Apr 24 21:21 tstat.out
-rw-r--r-- 1 root system 129628 Apr 24 21:24 tunables.sum
-rw-r--r-- 1 root system 28685 Apr 24 21:24 tunables_lastboot
-rw-r--r-- 1 root system 1365 Apr 24 21:24 tunables_lastboot.log
-rw------- 1 root system 419 Apr 24 21:24 tunables_nextboot
-rw------- 1 root system 419 Apr 24 21:24 tunables_nextliveupdate
-rw------- 1 root system 791 Apr 24 21:24 tunables_usermodified
-rw-r--r-- 1 root system 26452 Apr 24 21:24 tunablesx.sum
-rw-r--r-- 1 root system 109272 Apr 24 21:25 unix.what
-rw-r--r-- 1 root system 13668 Apr 24 21:21 ventstat.out
-rw-r--r-- 1 root system 6001 Apr 24 21:25 vfc_client.kdb.out
-rw-r--r-- 1 root system 1423 Apr 24 21:25 vfs.kdb
-rw-r--r-- 1 root system 29712 Apr 24 21:21 vmmstats_kdb.after
-rw-r--r-- 1 root system 29712 Apr 24 21:11 vmmstats_kdb.before
-rw-r--r-- 1 root system 2521 Apr 24 21:24 vmpools.out
-rw-r--r-- 1 root system 560 Apr 24 21:24 vmpools.save
-rw-r--r-- 1 root system 100743 Apr 24 21:21 vmstat.int
-rw-r--r-- 1 root system 465224 Apr 24 21:21 vmstat.psize.int
-rw-r--r-- 1 root system 1163 Apr 24 21:21 vmstat.sum
-rw-r--r-- 1 root system 2112 Apr 24 21:21 vmstat_s.out
-rw-r--r-- 1 root system 2475 Apr 24 21:21 vmstat_s.p.after
-rw-r--r-- 1 root system 2475 Apr 24 21:11 vmstat_s.p.before
-rw-r--r-- 1 root system 1083 Apr 24 21:21 vmstat_v.after
-rw-r--r-- 1 root system 1083 Apr 24 21:11 vmstat_v.before
-rw-r--r-- 1 root system 1802 Apr 24 21:21 vmstati.after
-rw-r--r-- 1 root system 1802 Apr 24 21:11 vmstati.before
-rw-r--r-- 1 root system 135 Apr 24 21:25 vnic.out
-rw-r--r-- 1 root system 135 Apr 24 21:21 vnicstats.after
-rw-r--r-- 1 root system 135 Apr 24 21:11 vnicstats.before
-rw-r--r-- 1 root system 4496747 Apr 24 21:25 vnode.kdb
-rw-r--r-- 1 root system 0 Apr 24 21:25 vxfs_tunables.out
-rw-r--r-- 1 root system 497 Apr 24 21:28 w.int
HACMP / PowerHA 利用時の注意
HACMP が稼働しているアクティブなクラスタノードで perfpmr を実行する場合、追加負荷による誤フェイルオーバーが発生する可能性があります。
特に Dead Man Switch モニタと同一ディスクにデータ収集する場合 は、事前にタイムアウト値を延長するか、別ディスクを使用してください。実行前に AIX HA チームへの確認を推奨します。
VIOS での収集
仮想デバイスが構成されている環境では、LPAR クライアントと 同時に VIOS 側の perfpmr データも収集することが推奨されます。
# root 権限を取得
$ oem_setup_env
# mkdir /tmp/perfdata
# cd /tmp/perfdata
# /opt/IBM/perfpmr/perfpmr.sh
まとめ
AIX 7.3 TL4 における perfpmr の標準統合により：
- 即時利用可能：ダウンロード・展開作業が不要
- 自動化が簡素化：固定パスで直接コール可能
- バージョン整合性：AIX 更新経由で常に最新状態
- 正式サポート：OS レベルで公式サポート対象
性能問題発生時に即座にデータ収集がよりスムーズにできるようになりました。
以上です。