Bug 1845266

Summary:	Significant slowdown in running jobs in PSI upshift - internal zuul (DNS configuration issue)
Product:	Red Hat OpenStack	Reporter:	Ronelle Landy <rlandy>
Component:	openstack-glance	Assignee:	Erno Kuvaja <ekuvaja>
Status:	CLOSED NOTABUG	QA Contact:	Mike Abrams <mabrams>
Severity:	high	Docs Contact:	Chuck Copello <ccopello>
Priority:	high
Version:	17.0 (Wallaby)	CC:	akekane, apevec, athomas, cyril, dhill, eglynn, ekuvaja, jmelvin, jvisser, kforde, mabrams, mwitt, pgrist, pweeks
Target Milestone:	---	Keywords:	Triaged
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2020-07-22 18:42:14 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 Ronelle Landy 2020-06-08 19:13:49 UTC

Description of problem:

Jobs running in the tripleo-ci-internal tenant have required twice the time to run to completion since late 06/03. The issue is tracked in https://access.redhat.com/support/cases/#/case/02671591.

Creating this BZ to log an associated CIX.

Comment 16 Alan Pevec 2020-07-07 20:08:02 UTC

Moving to Storage/Glance to analyze sosreports attached to the case.

Comment 29 Paul Grist 2020-07-21 13:07:09 UTC

I propose closing this bug as suggested and starting a new bug with a fresh context for the PSI team seeing very slow API response times.  Comments here and the case indicate that the CI slowdown originally escalated was resolved with the efforts here from Melanie.  In general, we need separate cases and bugs for new issues so we can track specific issues to closure including future prevention. 

Can someone elaborate on the impact being seen on the slow API times and can someone open a fresh bug?

I plan to close this notabug, but will give it a day to see if there's more feedback. The CIX card has been closed for a while now.

Comment 31 Paul Grist 2020-07-22 18:42:14 UTC

Closing this bug based on the feedback that the original specific bug and CiX issue was resolved by the DNS convfiguration changed and updated title to better indicate the issue this tracked.  Also if anyone feels we could to more to document or prevent this situation, please open a doc BZ based on what we learned to close the loop on this one.

Please open a new bug for the slow API performance if that is still being seen and investigated.

Thanks,
Paul