Bug 1120903

Summary:	[RFE] pulp worker count needs a maximum
Product:	Red Hat Satellite	Reporter:	Justin Sherrill <jsherril>
Component:	Installation	Assignee:	Eric Helms <ehelms>
Status:	CLOSED ERRATA	QA Contact:	Elyézer Rezende <erezende>
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	6.0.3	CC:	bbuckingham, bkearney, ehelms, erezende, jmontleo
Target Milestone:	Unspecified	Keywords:	FutureFeature
Target Release:	Unused
Hardware:	Unspecified
OS:	Unspecified
URL:	http://projects.theforeman.org/issues/8266
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-08-12 05:10:42 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:
Embargoed:

Description Justin Sherrill 2014-07-18 00:45:48 UTC

Description of problem:

currently if you install a satellite on systems with a large number of cpus (16, 32, etc..) the number of pulp workers is very large.  In reality we only need around 8 workers max.  So we should default the config file in /etc/defaults/ to 6 or 8 or the cpu count if its lower than 6 or 8

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


How reproducible:


Steps to Reproduce:
1.  Install satellite 6
2.  Check the number of workers  ps aux | grep pulp_workers (then divide by 2)


Actual results:
Does not limit itself to any number.

Expected results:
For a large cpu system should limit itself to 6 or 8

Additional info:

Comment 2 Eric Helms 2014-11-04 17:19:32 UTC

Created redmine issue http://projects.theforeman.org/issues/8266 from this bug

Comment 3 Bryan Kearney 2015-02-21 03:02:32 UTC

Moving to POST since upstream bug http://projects.theforeman.org/issues/8266 has been closed

Comment 6 Elyézer Rezende 2015-04-07 12:31:23 UTC

Verified on Satellite-6.1.0-RHEL-7-20150406.0

Steps to verify:

1. Installed Satellite
2. Fetched the CPU number
3. Checked the number of pulp workers (decreased 1 because grep matches itself)


Machine 1
=========

# lscpu | grep '^CPU(s):'
CPU(s):                32

# expr $(ps aux | grep -E "reserved_resource_worker-" | wc -l) - 1
16


Machine 2
=========

# lscpu | grep '^CPU(s):'
CPU(s):                24

# expr $(ps aux | grep -E "reserved_resource_worker-" | wc -l) - 1
16


To get the real number of pulp workers, the number should be divided by 2 as stated in the bug description, in both cases was 8.

Comment 7 Bryan Kearney 2015-08-11 13:34:24 UTC

This bug is slated to be released with Satellite 6.1.

Comment 8 errata-xmlrpc 2015-08-12 05:10:42 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2015:1592