Bug 2204523

Summary: ovs-vswitchd not creating DP flows
Product: Red Hat Enterprise Linux Fast Datapath Reporter: Timothy Redaelli <tredaelli>
Component: openvswitch3.1Assignee: Adrián Moreno <amorenoz>
Status: CLOSED ERRATA QA Contact: ovs-qe
Severity: unspecified Docs Contact:
Priority: high    
Version: RHEL 9.0CC: amorenoz, ctrautma, fleitner, i.maximets, jhsiao, ovs-qe, pdiak, ralongi, tredaelli
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: openvswitch3.1-3.1.0-10.el9fdp Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 2180460 Environment:
Last Closed: 2023-07-06 19:17:49 UTC Type: ---
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: 2180460    
Bug Blocks:    

Description Timothy Redaelli 2023-05-15 22:46:36 UTC 
+++ This bug was initially created as a clone of Bug #2180460 +++

OVN-Kubernetes issue: https://issues.redhat.com/browse/OCPBUGS-10485

Description of problem:
During OCP node startup network setup fails because DHCP packets are not leaving the node. 

It looks like there are no DP flows programmed:
ovs-appctl dpctl/show
system@ovs-system:
  lookups: hit:0 missed:620282 lost:620282
  flows: 0
  masks: hit:0 total:0 hit/pkt:0.00
  caches:
    masks-cache: size:256
  port 0: ovs-system (internal)
  port 1: ens4
  port 2: ovn-k8s-mp0 (internal)
  port 3: genev_sys_6081 (geneve: packet_type=ptap)
  port 4: br-int (internal)
  port 5: br-ex (internal)

ovs-ofctl dump-flows br-ex
 cookie=0x0, duration=171.388s, table=0, n_packets=5, n_bytes=210, priority=0 actions=NORMAL

ovs-appctl dpctl/dump-flows
<NO_OUTPUT>


Version-Release number of selected component (if applicable):
ovs_version: "3.1.1"
OCP version: 4.13.0-0.nightly-2023-03-14-053612

How reproducible: Often, but only when using nodes with two cores


Steps to Reproduce:
Setup OCP 4.13.0-0.nightly-2023-03-14-053612 using workers that have 2 CPU cores each

Actual results:
ovs-configuration.service fails to setup br-ex, DHCP traffic not leaving the node

Expected results:
network setup succedes

Additional info:
Looking at the attached ovs-vswitchd it looks like there are initially 0 n-handler-threads configured

--- Additional comment from Patryk Diak on 2023-03-21 18:06:29 CET ---

slack thread: https://redhat-internal.slack.com/archives/C04L7QWC9CZ/p1679318992314329

--- Additional comment from Patryk Diak on 2023-03-22 12:08:47 CET ---



--- Additional comment from Patryk Diak on 2023-03-22 12:49:39 CET ---



--- Additional comment from Patryk Diak on 2023-03-22 12:50:35 CET ---



--- Additional comment from Ilya Maximets on 2023-03-28 15:04:59 CEST ---

Automation failed to update the BZ, so doing that manually.

Timothy, this one will need some manual cloning, I suppose.
Comment 3 Christian Trautman 2023-06-07 15:25:47 UTC 
This was tested as part of OCP CI per https://issues.redhat.com/browse/OCPBUGS-10485
Comment 7 errata-xmlrpc 2023-07-06 19:17:49 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 (openvswitch3.1 bug fix and enhancement update), 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/RHBA-2023:3989