https://aznot.com/index.php?action=history&feed=atom&title=OpenWest_2014%2FLinux_Performance_Tools 2026-04-30T05:52:37Z Revision history for this page on the wiki MediaWiki 1.41.0 https://aznot.com/index.php?title=OpenWest_2014/Linux_Performance_Tools&diff=48&oldid=prev 2014-05-11T15:28:05Z

<p></p> <p><b>New page</b></p><div>by Ed Felt - DBA/Developer at LDS Church<br /> <br /> Keep it simple ... at least at first<br /> <br /> No Mockumentation - Document well, even over document<br /> <br /> Recommended course - Linux Foundation - Linux Performance Tools (LF426) - ~$3,000<br /> * https://training.linuxfoundation.org/linux-courses/system-administration-training/linux-performance-tuning<br /> <br /> Track changes as you test<br /> <br /> Human memory sucks - keep personal notes (journaling)<br /> <br /> Keep focused on one variable at a time (Exception: fix it now emergency fire!) - most of us think linearly<br /> <br /> Examples:<br /> * ngrep to troubleshoot Segmentation Fault that doesn&#039;t make it to the logs<br /> * actime0=0 # NFS (?kill it/keep it?)<br /> <br /> Stress testing is NOT base-lining/profiling or even troubleshooting<br /> <br /> /proc - TLDP describes it as &quot;window&quot; in to kernel and processes<br /> <br /> PIDS:<br /> * /proc/PID/cmdline<br /> * /proc/PID/cwd<br /> * /proc/PID/environ<br /> * /proc/PID/exe<br /> * ps<br /> <br /> CPU:<br /> * /proc/cpuinfo<br /> * top<br /> <br /> Memory:<br /> * /proc/meminfo<br /> * free<br /> <br /> Netstat:<br /> * /proc/net/netstat<br /> * netstat<br /> <br /> Logs:<br /> * /var/log/messages<br /> * /var/log/[application]<br /> <br /> Write your own:<br /> * BASH<br /> * Perl<br /> * Python<br /> <br /> Tools:<br /> * hdparm -tT /dev/....<br /> * top<br /> * sar (sysstat)<br /> * collectl<br /> * mlmon<br /> <br /> Other tools:<br /> * watch<br /> * tail<br /> * strace<br /> * iostat<br /> * vmstat<br /> * nmon<br /> * iptraf<br /> * lsof<br /> * htop</div>

Kenneth