Timeline for 100.0%sy with poor disk performance
Current License: CC BY-SA 3.0
22 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 26, 2012 at 13:04 | comment | added | phemmer | Thanks. Wasnt familiar with that VM setting. Will definitely keep it noted. | |
| Jan 26, 2012 at 8:48 | history | edited | user14199 | CC BY-SA 3.0 |
deleted 764 characters in body
|
| Jan 26, 2012 at 8:47 | answer | added | user14199 | timeline score: 3 | |
| Jan 26, 2012 at 8:46 | comment | added | user14199 | @Patrick we found the problem vm.zone_reclaim_mode was set by default to 1 we disabled it by sysctl vm.zone_reclaim_mode=0 and its not happened again | |
| Jan 26, 2012 at 8:41 | history | edited | user14199 | CC BY-SA 3.0 |
added 764 characters in body
|
| Jan 17, 2012 at 7:52 | comment | added | user14199 | yes that what we did. i will update here with the progress | |
| Jan 16, 2012 at 23:48 | comment | added | phemmer | ya, i'm clueless. something is causing the kernel to spend way too long during write() operations it seems. I would open an offical support ticket with redhat at this point. | |
| Jan 16, 2012 at 12:59 | comment | added | user14199 | @Patrick /dev/mapper/VolGroup00-LogVol00 on / type ext3 (rw) | |
| Jan 16, 2012 at 12:38 | comment | added | phemmer |
Thats whats in fstab, what does mount say?
|
|
| Jan 16, 2012 at 8:08 | comment | added | user14199 | yes it without it / ext3 defaults 1 1 | |
| Jan 16, 2012 at 5:27 | history | tweeted | twitter.com/#!/StackUnix/status/158782475336814592 | ||
| Jan 15, 2012 at 17:17 | comment | added | phemmer |
The filesystem isnt mounted with sync turned on is it?
|
|
| Jan 15, 2012 at 17:01 | comment | added | phemmer | I deleted my answer as its obviously incorrect based on the info you provided. This shows that its not interrupts or anything, but really is the kernel consuming all the CPU time. I cannot think of why this would be. The only thing I can think of is something wrong with the filesystem (and so filesystem driver is eating CPU). You might also try upgrading to one of the later RHEL5 point releases (5.7 is the latest). | |
| Jan 15, 2012 at 9:00 | comment | added | rvs | redhat ports some features as well | |
| Jan 15, 2012 at 8:40 | history | edited | user14199 | CC BY-SA 3.0 |
added 5540 characters in body
|
| Jan 14, 2012 at 2:36 | comment | added | phemmer | @psusi redhat backports critical fixes to older kernels. It may be 2.6.18, but its still up to date with fixes. The only thing missing vs newer kernels would be features. | |
| Jan 12, 2012 at 14:59 | comment | added | Karlson | @psusi Given the user running RHEL and Oracle that may not be an option... | |
| Jan 11, 2012 at 15:18 | comment | added | psusi | 2.6.18 is an ancient kernel ( it was released in 2006! ). You really should upgrade. | |
| Jan 11, 2012 at 12:17 | answer | added | symcbean | timeline score: 0 | |
| Jan 11, 2012 at 12:04 | history | edited | Mat | CC BY-SA 3.0 |
added 55 characters in body
|
| Jan 11, 2012 at 11:39 | history | edited | user14199 | CC BY-SA 3.0 |
added 172 characters in body
|
| Jan 11, 2012 at 11:30 | history | asked | user14199 | CC BY-SA 3.0 |