You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Also we have enabled audit logs in vault and configured logrotate config to rotate the logs on daily basis. We also have added a command to reload it as mentioned in the docs (basically sending a HUP signal to the process)
The issue here is with systemctl reload vault . We have tested it with different versions of vault and it is deregistering the consul service starting v1.17.0 version when it is working properly in v1.16.3. when I reload vault in v1.17.0 or later I get this in the logs:
[DEBUG] service_registration.consul: registration is now empty, deregistering service from consul
The consul dns entry for the vault cluster is used widely by the other apps and is a very important component in our installation, but we lose it after logrotate proceeds with the rotation.
To fix this I have to restart the service and unseal the nodes to get the consul service registered again.
I didn't find anything in release notes or issues regarding this specific issue. Are we doing something wrong or is this indeed a bug?
Thank you.
The text was updated successfully, but these errors were encountered:
We have a vault cluster running the latest v1.18.3 which is using consul as storage and also registering the service in consul:
Also we have enabled audit logs in vault and configured logrotate config to rotate the logs on daily basis. We also have added a command to reload it as mentioned in the docs (basically sending a HUP signal to the process)
The issue here is with
systemctl reload vault
. We have tested it with different versions of vault and it is deregistering the consul service starting v1.17.0 version when it is working properly in v1.16.3. when I reload vault in v1.17.0 or later I get this in the logs:[DEBUG] service_registration.consul: registration is now empty, deregistering service from consul
The consul dns entry for the vault cluster is used widely by the other apps and is a very important component in our installation, but we lose it after logrotate proceeds with the rotation.
To fix this I have to restart the service and unseal the nodes to get the consul service registered again.
I didn't find anything in release notes or issues regarding this specific issue. Are we doing something wrong or is this indeed a bug?
Thank you.
The text was updated successfully, but these errors were encountered: