Comments on: Reducing inode and dentry caches to keep OOM killer at bay http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/ Words of wisdom from a server administrator Sat, 11 Feb 2012 23:43:21 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Liberando la memoria Cache | El atareao http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-23929 Liberando la memoria Cache | El atareao Wed, 17 Aug 2011 22:02:18 +0000 http://rackerhacker.com/?p=635#comment-23929 [...] embargo, y después de leer Racker Hacker, sobre este asunto, me quedé incluso más confundido. En este punto se comenta de una posibilidad [...] [...] embargo, y después de leer Racker Hacker, sobre este asunto, me quedé incluso más confundido. En este punto se comenta de una posibilidad [...]

]]> By: Seth http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-23425 Seth Thu, 07 Jul 2011 16:13:19 +0000 http://rackerhacker.com/?p=635#comment-23425 > That page says "If you are writing data at the time you run these commands, you'll actually be dumping the data out of the filesystem cache before it reaches the disk, which could lead to very bad things". Is decidedly NOT TRUE. Andrew Morton himself (!!!) took the trouble to comment as much. Would you please fix your statement? @linux_user: Doing a synch first changes nothing, because other processes can (and will) be dirtying page cache in the mean time > That page says "If you are writing data at the time you run these
commands, you'll actually be dumping the data out of the filesystem
cache before it reaches the disk, which could lead to very bad things".

Is decidedly NOT TRUE. Andrew Morton himself (!!!) took the trouble to comment as much. Would you please fix your statement?

@linux_user:
Doing a synch first changes nothing, because other processes can (and will) be dirtying page cache in the mean time

]]> By: Alex http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-16456 Alex Tue, 03 Aug 2010 14:17:09 +0000 http://rackerhacker.com/?p=635#comment-16456 Thank you for this article. I was banging my head for long time with VMs web servers caching 60%+ of the available memory and becoming slow as a turtle! I am referencing your article in my resent "research", setting up <a href="http://boolcast.com/lamp/linux/7-centos-5-minimalist" rel="nofollow">Cent OS 5 Minimalist</a> Thank you for this article. I was banging my head for long time with VMs web servers caching 60%+ of the available memory and becoming slow as a turtle!
I am referencing your article in my resent "research", setting up Cent OS 5 Minimalist

]]> By: Alex http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-16231 Alex Tue, 11 May 2010 08:01:29 +0000 http://rackerhacker.com/?p=635#comment-16231 From: Andrew Morton That page says "If you are writing data at the time you run these commands, you'll actually be dumping the data out of the filesystem cache before it reaches the disk, which could lead to very bad things". That had better not be true! That would be a bad bug. drop_caches only drops stuff which has been written back. http://lkml.indiana.edu/hypermail/linux/kernel/1005.1/00693.html From: Andrew Morton
That page says "If you are writing data at the time you run these
commands, you'll actually be dumping the data out of the filesystem
cache before it reaches the disk, which could lead to very bad things".

That had better not be true! That would be a bad bug. drop_caches
only drops stuff which has been written back.
http://lkml.indiana.edu/hypermail/linux/kernel/1005.1/00693.html

]]> By: linux_user http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-15630 linux_user Mon, 07 Dec 2009 01:42:45 +0000 http://rackerhacker.com/?p=635#comment-15630 Isn't it safe to do sync; and then, echo 3 > /proc/sys/vm/drop_caches instead of the other way around? Isn't it safe to do

sync; and then,
echo 3 > /proc/sys/vm/drop_caches

instead of the other way around?

]]> By: Oracle DBA Help » Truth about drop_cache & sync and dealing with OOM Killer http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-15380 Oracle DBA Help » Truth about drop_cache & sync and dealing with OOM Killer Tue, 30 Jun 2009 01:55:34 +0000 http://rackerhacker.com/?p=635#comment-15380 [...] day, while googling about a memory issue, I ended up on this page. It describes how to force discard some memory resident objects by writing 1, 2 or 3 into [...] [...] day, while googling about a memory issue, I ended up on this page. It describes how to force discard some memory resident objects by writing 1, 2 or 3 into [...]

]]> By: Savage http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-5418 Savage Sat, 06 Dec 2008 07:00:55 +0000 http://rackerhacker.com/?p=635#comment-5418 This should lead to another great post regarding the way oom works. As well I normally find nfs_inode_cache another good one to find. It is not to always perform a sync ; drop cache because you think this may be the cause. You can find out if it is due to this by cating /proc/slab or running slabtop. Now how do you keep oom_killer at bay with a 2.6.X kernel and not kill your services that is essential to the server if your sure that said service is not the problem you can set oom_adj in /proc//oom_adj this will raise / lower the priory of it's chance of being killed and you can then slowly build a way to control oom_killer. After this is done you can then check the score of your application which is in /proc//oom_score this should give you the current score and you can check your applications score at that time and find which may be killed first. Kernel Documentation : Documentation/filesystem/proc.txt " 2.12 /proc//oom_adj - Adjust the oom-killer score ------------------------------------------------------ This file can be used to adjust the score used to select which processes should be killed in an out-of-memory situation. Giving it a high score will increase the likelihood of this process being killed by the oom-killer. Valid values are in the range -16 to +15, plus the special value -17, which disables oom-killing altogether for this process. 2.13 /proc//oom_score - Display current oom-killer score ------------------------------------------------------------- ------------------------------------------------------------------------------ This file can be used to check the current score used by the oom-killer is for any given . Use it together with /proc//oom_adj to tune which process should be killed in an out-of-memory situation. " This should lead to another great post regarding the way oom works. As well I normally find nfs_inode_cache another good one to find. It is not to always perform a sync ; drop cache because you think this may be the cause. You can find out if it is due to this by cating /proc/slab or running slabtop.

Now how do you keep oom_killer at bay with a 2.6.X kernel and not kill your services that is essential to the server if your sure that said service is not the problem you can set oom_adj in /proc//oom_adj this will raise / lower the priory of it's chance of being killed and you can then slowly build a way to control oom_killer. After this is done you can then check the score of your application which is in /proc//oom_score this should give you the current score and you can check your applications score at that time and find which may be killed first.

Kernel Documentation :

Documentation/filesystem/proc.txt

"
2.12 /proc//oom_adj - Adjust the oom-killer score
------------------------------------------------------

This file can be used to adjust the score used to select which processes
should be killed in an out-of-memory situation. Giving it a high score will
increase the likelihood of this process being killed by the oom-killer. Valid
values are in the range -16 to +15, plus the special value -17, which disables
oom-killing altogether for this process.

2.13 /proc//oom_score - Display current oom-killer score
-------------------------------------------------------------

------------------------------------------------------------------------------
This file can be used to check the current score used by the oom-killer is for
any given . Use it together with /proc//oom_adj to tune which
process should be killed in an out-of-memory situation.
"

]]> By: major http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-5404 major Fri, 05 Dec 2008 23:20:18 +0000 http://rackerhacker.com/?p=635#comment-5404 I'm glad it was helpful! I've found some other optimizations as well, so I'll be adding those soon. I'm glad it was helpful! I've found some other optimizations as well, so I'll be adding those soon.

]]> By: Ace http://rackerhacker.com/2008/12/03/reducing-inode-and-dentry-caches-to-keep-oom-killer-at-bay/#comment-5403 Ace Fri, 05 Dec 2008 23:06:18 +0000 http://rackerhacker.com/?p=635#comment-5403 Thank you for this post, it was very very helpful. Thank you for this post, it was very very helpful.

]]>