Bug 1019142 - oo-accept-node show error messages when restarting node server or creating apps in parallel
Summary: oo-accept-node show error messages when restarting node server or creating ap...
Keywords:
Status: CLOSED DEFERRED
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Containers
Version: 2.0.0
Hardware: Unspecified
OS: Unspecified
medium
medium
Target Milestone: ---
: ---
Assignee: Brenton Leanhardt
QA Contact: libra bugs
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2013-10-15 08:27 UTC by Ma xiaoqiang
Modified: 2017-03-08 17:36 UTC (History)
3 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2014-01-21 13:13:25 UTC
Target Upstream Version:


Attachments (Terms of Use)

Description Ma xiaoqiang 2013-10-15 08:27:45 UTC
Description of problem:
oo-accept-node show error messages when restarting node server or creating apps in parallel

Version-Release number of selected component (if applicable):
puddle:[2.0/2013-10-14.3] 
http://buildvm-devops.usersys.redhat.com/puddle/build/OpenShiftEnterpriseErrata/2.0/2013-10-14.3/

How reproducible:
always

Steps to Reproduce:
first method:
1.create several apps
2.reboot the node server
3.run "oo-accept-node"
#oo-accept-node
second method:
1.create several app in parallel
for i in `seq 1 9`;do rhc app create testapp$i php & done
2.run "oo-accept-node"
#oo-accept-node


Actual results:
The similar errors are given out at random

Output:
FAIL: 525ce8455f101a5da90002f4 has a process missing from cgroups: 9385 cgroups controller: freezer
FAIL: 525ce8455f101a5da90002f4 has a process missing from cgroups: 9393 cgroups controller: freezer
FAIL: 525ce84c5f101a395f0019a1 has a process missing from cgroups: 10102 cgroups controller: net_cls
FAIL: 525ce84c5f101a395f0019a1 has a process missing from cgroups: 10106 cgroups controller: net_cls
FAIL: 525ce84c5f101a395f0019a1 has a process missing from cgroups: 10102 cgroups controller: freezer
FAIL: 525ce84c5f101a395f0019a1 has a process missing from cgroups: 10106 cgroups controller: freezer

That means some processes are not limited by cgroup.

Expected results:
Should show no error!! 

Additional info:
If the above error message is shown, restart cgred service, re-run oo-accept-node, it works fine.

Comment 3 Ma xiaoqiang 2014-01-21 04:27:25 UTC
I check this problem on devenv__4247 and puddle [2.0.2/2014-01-16.1], Now the problem can not be reproduced.
# ll /etc/rc.d/rc3.d/S[0-9][0-9]c*
lrwxrwxrwx. 1 root root 18 Jan 14 23:14 /etc/rc.d/rc3.d/S05cgconfig -> ../init.d/cgconfig
lrwxrwxrwx. 1 root root 18 Jan 14 23:03 /etc/rc.d/rc3.d/S13cpuspeed -> ../init.d/cpuspeed
lrwxrwxrwx. 1 root root 14 Jan 14 23:01 /etc/rc.d/rc3.d/S25cups -> ../init.d/cups
lrwxrwxrwx. 1 root root 15 Jan 14 23:14 /etc/rc.d/rc3.d/S30cgred -> ../init.d/cgred
lrwxrwxrwx. 1 root root 15 Jan 14 23:01 /etc/rc.d/rc3.d/S90crond -> ../init.d/crond
lrwxrwxrwx. 1 root root 20 Mar 21  2013 /etc/rc.d/rc3.d/S99certmonger -> ../init.d/certmonger

The cgred service has been started after the cgconfig service. So the problem can not be reproduced.


Note You need to log in before you can comment on or make changes to this bug.