Comment 4 for bug 1797297

Revision history for this message
Haw Loeung (hloeung) wrote : Re: leader_id stale/incorrect

To add:

2020-12-17 16:17:45 INFO juju.worker.leadership tracker.go:194 ubuntu-repository-cache/1 promoted to leadership of ubuntu-repository-cache
2020-12-17 16:18:07 INFO juju-log Reactive main running for hook leader-elected
tracer: ++ queue handler reactive/ubuntu-repository-cache.py:218:leader_elected
2020-12-17 16:18:07 INFO juju-log Invoking reactive handler: reactive/ubuntu-repository-cache.py:218:leader_elected
2020-12-17 16:18:07 INFO juju-log leader-elected fired. This is not the leader
2020-12-17 16:18:07 INFO juju.worker.uniter.operation runhook.go:142 ran "leader-elected" hook (via explicit, bespoke hook script)
2020-12-17 22:28:25 INFO juju.worker.leadership tracker.go:194 ubuntu-repository-cache/1 promoted to leadership of ubuntu-repository-cache
2020-12-17 22:48:09 INFO juju-log cluster:1: Updating metadata on the leader
2020-12-17 22:48:09 WARNING juju-log cluster:1: Leader changed between peer_update_metadata and _leader_update_metadata

And:

| ubuntu-repository-cache/1* unknown idle 1 20.195.53.225 80/tcp

However:

| ubuntu@machine-1:~$ sudo juju-run ubuntu-repository-cache/1 "leader-get"
| leader_id: ubuntu-repository-cache/0