Bug 2015504

Summary: disruption to ingress-to-console new connections on gcp-ovn
Product: OpenShift Container Platform Reporter: Devan Goodwin <dgoodwin>
Component: NetworkingAssignee: jamo luhrsen <jluhrsen>
Networking sub component: ovn-kubernetes QA Contact: Anurag saxena <anusaxen>
Status: CLOSED DUPLICATE Docs Contact:
Severity: high    
Priority: medium CC: astoycos
Version: 4.10   
Target Milestone: ---   
Target Release: 4.10.0   
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: 2021-11-05 15:41:01 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 Devan Goodwin 2021-10-19 11:44:16 UTC 
Disruption data is showing an outlier up around 50s mean for new connections in this combination.

https://datastudio.google.com/reporting/26007bde-88c8-44dd-8fb9-4f5e41dfb0ce/page/p_i6fa5cr1mc

Select Release 4.10, Network ovn, and scroll down to see Mean Disruption ingress-to-console new. gcp is showing around 50s for new connections.

For comparison other platforms on ovn are showing <10s.

For re-used connections and gcp, this is also higher than the others at 20s. (compared to <2s)

TRT is rating at high severity as this is a strong indicator that customer workloads will suffer the same kind of outage during an upgrade.
Comment 1 jamo luhrsen 2021-11-05 15:41:01 UTC 
marking this a duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=2015512 for now. preliminary tests
with this small fix in ovnk seems like it is helping:
  https://github.com/openshift/ovn-kubernetes/pull/806

*** This bug has been marked as a duplicate of bug 2015512 ***