Force recreation of ovs db cluster on first replica when protocol changes #361
Conversation
olliewalsh
requested review from
karelyatin and
stuggi
and removed request for
lewisdenny and
frenzyfriday
olliewalsh
force-pushed
the
tls_change
branch
from
37f6339 to
f493261
Compare
|
/hold |
…nges Cannot remove final cluster member to change protocol, but can convert to standalone schema. Cluster gets recreated with the correct address.
olliewalsh
force-pushed
the
tls_change
branch
from
d56257c to
2db8f74
Compare
|
/test ovn-operator-build-deploy-kuttl |
1 similar comment
|
/test ovn-operator-build-deploy-kuttl |
The liveness probe is quite agressive and could trigger before ovsdb-server starts when toggling TLS. Add a startup probe to tolerate this.
olliewalsh
force-pushed
the
tls_change
branch
from
9f6eb3c to
78eb4b2
Compare
|
/hold |
Deleting clusters when relicas=3 does not work reliably, eventually getting killed when hitting the termination grace period timeout. ovn_db_delete and ovn_tls_enable kuttl test are affected by this so for now just increase the timeout.
olliewalsh
force-pushed
the
tls_change
branch
from
78eb4b2 to
8d760a2
Compare
|
/remove-hold |
|
/test ovn-operator-build-deploy-kuttl |
1 similar comment
|
/test ovn-operator-build-deploy-kuttl |
| template='{{.status.internalDbAddress}}{{"\n"}}' | ||
| regex="tcp:.*" | ||
| dbUri=$(oc get -n $NAMESPACE OVNDBCluster ovndbcluster-sb-sample -o go-template="$template") | ||
| matches=$(echo "$dbUri" | sed -e "s?$regex??") | ||
| if [[ -n "$matches" ]]; then | ||
| exit 1 | ||
| fi |
There was a problem hiding this comment.
would this be the same as ?
oc get -n $NAMESPACE OVNDBCluster ovndbcluster-sb -o jsonpath={.status.internalDbAddress} | grep -q tcp
There was a problem hiding this comment.
Yes, would use grep -q '^tcp:'.
| # Check ovn connect is ptcp | ||
| - script: | | ||
| sb_pod=$(oc get pod -n $NAMESPACE -l service=ovsdbserver-sb -o name|head -1) | ||
| oc rsh -n $NAMESPACE ${sb_pod} ovn-sbctl --no-leader-only get-connection | grep -q ptcp |
There was a problem hiding this comment.
can use oc rsh or oc exec on the deployment to not have to check for the pod I think
There was a problem hiding this comment.
yes, seems to work
$ oc rsh statefulset/ovsdbserver-sb hostname
ovsdbserver-sb-0
This and the other assert were taken from the existing kutt tests so maybe clean them all up in a follow up patch?
|
@otherwiseguy: changing LGTM is restricted to collaborators In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: karelyatin, olliewalsh, otherwiseguy The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-merge-bot
bot
merged commit b2f1022
into
openstack-k8s-operators:main
Cannot remove final cluster member to change protocol, but can convert to standalone schema. Cluster gets recreated with the correct address.
Jira: OSPRH-6954
Jira: OSPRH-10415