Bug 1523064

Summary:	Potential memory leak in neutron-server
Product:	Red Hat OpenStack	Reporter:	Meiyan Zheng <mzheng>
Component:	openstack-neutron-lbaas	Assignee:	Nir Magnezi <nmagnezi>
Status:	CLOSED CURRENTRELEASE	QA Contact:	Bruna Bonguardo <bbonguar>
Severity:	high	Docs Contact:
Priority:	high
Version:	10.0 (Newton)	CC:	amuller, apevec, astafeye, bbonguar, bcafarel, bhaley, cgoncalves, chrisw, dalvarez, jlibosva, lhh, majopela, mbayer, mzheng, pvarma, ragiman, skaplons, slinaber, sputhenp, srevivo
Target Milestone:	zstream	Keywords:	TestOnly, Triaged, ZStream
Target Release:	10.0 (Newton)
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:	openstack-neutron-lbaas-9.2.2-8.el7ost	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-06-07 18:02:22 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:
Bug Depends On:	1545939
Bug Blocks:

Comment 12 Praveen Varma 2018-08-15 01:03:31 UTC

Team,

There is nothing on this BZ for sometime now even though I can assume that there is work being done in the background.

There is an ACE that is created for the associated case and I am involved to see how we can take this forward.

Can I please get a status update on the BZ and if possible have this prioritized?

Thanks,
Praveen
Escalation Manager

Comment 41 Assaf Muller 2018-12-12 14:28:37 UTC

This is dependent on 1545939. The fix for that other bug was delivered as a hotfix to the customer impacted by this bug.

Comment 50 Bruna Bonguardo 2019-04-03 12:45:24 UTC

Nir, what are the exact steps to reproduce/verify this in my environment?

Comment 51 Nir Magnezi 2019-04-26 10:04:10 UTC

(In reply to Bruna Bonguardo from comment #50)
> Nir, what are the exact steps to reproduce/verify this in my environment?

Hi Bruna,

We spoke f2f about this, but it makes sense to keep the information here up-to-date:
I advised repeating the steps Slawek took on comment 39, which should give an indication for quality engineering to verify this.

If you need any help with this, please do let me know.

Nir

Comment 56 Bruna Bonguardo 2019-04-29 13:22:18 UTC

As you can see, I created 10 load balancers with 10 listeners each. The memory is as stated above. It does not look unreasonably high.

Nir - can you connect to the server with me and check?

Comment 67 Slawek Kaplonski 2019-05-20 06:52:58 UTC

Yes, this looks fine. Memory is allocated by python in first iteration and it's not given back to OS after resources are removed. But in next iterations same memory is reused by new resources thus consumption is not raising anymore.
I don't see any leak in https://docs.google.com/spreadsheets/d/147vG-PO8SAijjK4sR87K-7v3YGfR0fae5AlOdiXZwrA/edit#gid=0
IMO it works fine.

Comment 69 Jason Joyce 2019-06-07 18:02:22 UTC

According to our records, this should be resolved by openstack-neutron-lbaas-9.2.2-8.el7ost.  This build is available now.