-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
None
-
3
-
False
-
-
False
-
?
-
?
-
?
-
?
-
None
-
-
-
RHOSO sprint BugFix 4
-
Important
When `edpm_ovn_bgp_agent_local_ovn_routing` is enabled, some containers are deployed on each EDPM node, implementing an OVN cluster: NB DB, SB DB, northd and ovn-controller. See [1].
On RHOSO 18.0 FR1, `northd` and `bgp_ovn_controller` containers are properly running on EDPM nodes, but `nb_db_server` and `sb_db_server` are restarting every few seconds.
The reason is the options added by edpm-ansible to the `ovn-ctl` command executed on these containers when TLS is enabled are not valid:
Nov 11 18:22:06 compute-ehm6ti6v-0 nb_db_server[30040]: Running command: '/usr/share/ovn/scripts/ovn-ctl --no-monitor run_nb_ovsdb -p /etc/pki/tls/private/ovndb.key -c /etc/pki/tls/certs/ovndb.crt -C /etc/pki/tls/certs/ovndbca.crt ' Nov 11 18:22:06 compute-ehm6ti6v-0 nb_db_server[30040]: + echo 'Running command: '\''/usr/share/ovn/scripts/ovn-ctl --no-monitor run_nb_ovsdb -p /etc/pki/tls/private/ovndb.key -c /etc/pki/tls/certs/ovndb.crt -C /etc/pki/tls/certs/ovndbca.crt '\''' Nov 11 18:22:06 compute-ehm6ti6v-0 nb_db_server[30040]: /usr/share/ovn/scripts/ovn-ctl: unknown option "-p" (use --help for help) Nov 11 18:22:07 compute-ehm6ti6v-0 systemd[1]: edpm_nb_db_server.service: Main process exited, code=exited, status=1/FAILURE
Those options can be found here [2].
