Bug 1396052

Summary: OOM Killer randomly kills processes with 8.5 GB of free ram
Product: [Fedora] Fedora Reporter: customercare
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED EOL QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: urgent Docs Contact:
Priority: unspecified    
Version: 23CC: gansalmon, ichavero, itamar, jonathan, kernel-maint, madhu.chinakonda, mchehab, trevor
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-12-20 21:43:02 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
var-log-messages output of kernel-ooms
none
a munin plot of the ram usage over the last week none

Description customercare 2016-11-17 11:02:15 UTC 
Created attachment 1221522 [details]
var-log-messages output of kernel-ooms

Description of problem:

Running kernel: 4.7.10-100.fc23.i686+PAE ( Experienced it with 4.7.9-100 too)
RAM: 10 GB
Utilization: 1.5 GB
Free Ram at OOM: 8.5 GB

The Server in question is a high load webservices on a production system,
so a "pattern" is not emerging, because it's not predictable, who accesses it when & how. 

Daemons like clamd are notoriosly often killed, because they grow fast & live short, which is the prefered viticim of the oom-killer logic ( no clue why, as any old process can go "mad" with ram requests any time ).

oviosly often, the oom process is originated by mysqld. But as for patterns, thats it. You don't know when or why it happens.

A logfile is attached.


Version-Release number of selected component (if applicable):

kernel-headers-4.7.10-100.fc23.i686
kernel-PAE-4.8.8-100.fc23.i686
kernel-PAE-core-4.7.10-100.fc23.i686
kernel-PAE-core-4.7.9-100.fc23.i686
kernel-PAE-core-4.8.8-100.fc23.i686
kernel-PAE-modules-4.7.10-100.fc23.i686
kernel-PAE-modules-4.7.9-100.fc23.i686
kernel-PAE-modules-4.8.8-100.fc23.i686




How reproducible:

Random.

Actual results:

random OOM kills of partly important daemons.

Expected results:


Additional info:

Related to:

http://www.spinics.net/lists/linux-mm/msg113661.html
https://bugzilla.redhat.com/show_bug.cgi?id=1373339

Suggested Workaround by Bug #1373339 as a hourly cron : 

#!/bin/bash

sync && echo 1 > /proc/sys/vm/drop_caches

date -R
echo "Flushing of Caches done" >> /var/log/messages;
Comment 1 customercare 2016-11-17 11:03:51 UTC 
Created attachment 1221523 [details]
a munin plot of the ram usage over the last week

the break down in caches is a result of the reboot on monday switching from kernel 4.7.9 to 4.7.10 .
Comment 2 customercare 2016-11-17 12:59:42 UTC 
As the server farm used the same kernel, i checked for other servers with OOM and did not find any, but all others do not have more than 4 GB or near the activity that server has.
Comment 3 customercare 2016-11-17 16:34:09 UTC 
"sync && echo 1 > /proc/sys/vm/drop_caches"

Does not help under 4.7.10 :(
Comment 4 Fedora End Of Life 2016-11-25 09:40:33 UTC 
This message is a reminder that Fedora 23 is nearing its end of life.
Approximately 4 (four) weeks from now Fedora will stop maintaining
and issuing updates for Fedora 23. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as EOL if it remains open with a Fedora  'version'
of '23'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora 23 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.
Comment 5 Fedora End Of Life 2016-12-20 21:43:02 UTC 
Fedora 23 changed to end-of-life (EOL) status on 2016-12-20. Fedora 23 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.