Red Hat Satellite engineering is moving the tracking of its product development work on Satellite to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "Satellite project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs will be migrated starting at the end of May. If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "Satellite project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/SAT-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 1268258 - mongodb fails to start with numa interleave policy on RHEL7.1
Summary: mongodb fails to start with numa interleave policy on RHEL7.1
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Red Hat Satellite
Classification: Red Hat
Component: Packaging
Version: 6.1.1
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: Unspecified
Assignee: satellite6-bugs
QA Contact: Katello QA List
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-10-02 10:33 UTC by Pradeep Kumar Surisetty
Modified: 2017-08-11 12:30 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2017-08-11 12:30:25 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Description Pradeep Kumar Surisetty 2015-10-02 10:33:07 UTC
Description of problem:

mongodb recommends to interleave memory across all nodes for better performance.  As part of this exercise, i tried to interleave mongodb and check satellite performance. Mongodb fails to start with NUMA interleave. 

http://docs.mongodb.org/manual/administration/production-notes/


1) interleave across nodes for mongod: 

numactl --interleave=all /usr/bin/mongod -f /etc/mongodb.conf


NUMA is interleaving across nodes as shown below. 


http://pastebin.test.redhat.com/316482

2) Start mongod. It fails to start. 

systemctl start mongod.service. 

Strace:

http://pastebin.test.redhat.com/316531

Its an issue with systemd.   

3) Added init.d script for mongod. Still fails to start.  

http://pastebin.test.redhat.com/317072

Without interleave, mongod starts successfully. 


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

mongodb-2.4.9-3.el7sat.x86_64
katello-2.2.0.16-1.el7sat.noarch
pulp-server-2.6.0.15-1.el7sat.noarch

How reproducible:


Steps to Reproduce:
1.  numactl --interleave=all /usr/bin/mongod -f /etc/mongodb.conf    
2.   systemctl start mongod.service

Actual results:

Mongod fails to start. 

Expected results:

Mongod should start after numa interleave



Additional info:

Comment 2 Pradeep Kumar Surisetty 2016-01-20 12:02:46 UTC
Ran mongodb with numa interleave on centos/ 
Not sure if this is the recommended one. 

http://psuriset.com/2015/12/20/starting-mongodb-on-centos-with-numa-disabled

Comment 3 Eric Helms 2016-12-16 16:47:10 UTC
Can you shed some light on what you believe the issue is? i.e. Do you believe it the issue is in the version of MongoDB we ship? The packaging around the MongoDB that we ship? An issue with systemd in general?

Comment 4 Pradeep Kumar Surisetty 2017-04-24 11:24:08 UTC
i believe its with mongodb we ship.

Comment 5 Bryan Kearney 2017-08-11 12:30:25 UTC
This bz has been moved to an upstream issue. We will no longer be tracking this in the downstream. When the upstream issue is fixed, the next release of Satellite will include the fix. If this is a concern, please reach out to Bryan Kearney or Rich Jerrido. Thank you.


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