Review test setup to verify if the setup works as expected (NETVIRT-1006)

[NETVIRT-1031] Review controller-max-backoff and controller-inactivity-probe values Created: 28/Nov/17  Updated: 17/Sep/18  Resolved: 17/Sep/18

Status: Resolved
Project: netvirt
Component/s: General
Affects Version/s: None
Fix Version/s: None

Type: Sub-task Priority: Medium
Reporter: Sam Hague Assignee: Unassigned
Resolution: Done Votes: 0
Labels: csit:3node
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Relates
relates to OPNFLWPLUG-1007 Future (eventually) failed: addFlow -... Resolved

 Description   

Should the the two values are used to configure how the ovs nodes connect to ODL be changed from the defaults?

  • controller-max-backoff: max time between connection attempts from the nodes to ODL
  • controller-inactivity-probe: time between sending inactivity probes on connections

The defaults are 8s and 5s. If the inactivity-probe hits, then the switches disconnect from ODL. That cleans up the openflowplugin operational data for the node. That is then relayed to networking-odl that the node is down and networking-odl marks the node as inactive. When a node is down, networking-odl driver would mark port-binding as failed, so VM would enter ERROR state

 

 Comments   
Comment by Sam Hague [ 17/Sep/18 ]

https://git.opendaylight.org/gerrit/#/c/51763/

controller-inactivity-probe was bumped to 30s

Generated at Wed Feb 07 20:23:04 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.