Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1220154

Summary:	[scale] vdsm runs potential memory leak
Product:	Red Hat Enterprise Virtualization Manager	Reporter:	Eldad Marciano <emarcian>
Component:	vdsm	Assignee:	Yaniv Bronhaim <ybronhei>
Status:	CLOSED DUPLICATE	QA Contact:	Eldad Marciano <emarcian>
Severity:	high	Docs Contact:
Priority:	unspecified
Version:	3.5.1	CC:	bazulay, danken, ecohen, emarcian, gklein, lpeer, lsurette, oourfali, ybronhei, yeylon
Target Milestone:	---
Target Release:	3.5.3
Hardware:	x86_64
OS:	Linux
Whiteboard:	infra
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-05-14 13:08:30 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	Infra	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Eldad Marciano 2015-05-10 13:44:02 UTC

Description of problem:
potential leak level discovered during a CHO scale test monitor.
the leak is minor and readable by overall memory usage \ vdsm pid memory usage graph's.
~300mb leaking during the time.
further investigation required using advanced python profilers.


please find more detailed report + graph's at mojo page.

Version-Release number of selected component (if applicable):
vdsm-4.16.12.1-3.el6ev.x86_64 (vt14.3)

How reproducible:
100%

Steps to Reproduce:
1. run 50 vms or more for 48 hours (idle vms).
2. expect growing at the memory usage.


Actual results:
growing at the memory usage monitor

Expected results:
stable memory usage once the guests are up and running and do not perform any kind if activity.


Additional info:

Comment 3 Dan Kenigsberg 2015-05-11 15:07:09 UTC

This is most like a dup of bug 1158108, as the fixing patch is included only in v4.16.15.

The memory leaked by bug 1158108 is proportional to the number of network devices on the host, tap devices (=vNICs) included.

Would yo please retest with rhev-3.5.3 ?

Comment 4 Oved Ourfali 2015-05-12 07:46:04 UTC

(In reply to Dan Kenigsberg from comment #3)
> This is most like a dup of bug 1158108, as the fixing patch is included only
> in v4.16.15.
> 
> The memory leaked by bug 1158108 is proportional to the number of network
> devices on the host, tap devices (=vNICs) included.
> 
> Would yo please retest with rhev-3.5.3 ?

Dan - The patch is merged. Shall I move it to MODIFIED, or you moved it to POST for a specific reason?

Comment 5 Eldad Marciano 2015-05-14 08:06:48 UTC

patch 40134 applied on 3.5 setup. 
i ran a 8 hours test, ans it still not clear if the patch fix the leak.

next step: 
run 48h CHO tets.