Closed fkondej closed 2 years ago
This is a segv because of prometheus metrics. Are we using the “Auction duration” metrics at all? It not removing it will fix this.
yes, we have metrics enabled: https://github.com/vegaprotocol/ansible/blob/master/roles/barenode/templates/vega/config.toml.j2#L122-L127
[Metrics]
Level = "Info"
Timeout = "5s"
Port = 2112
Path = "/metrics"
Enabled = true
What I meant was do we have any utility for this specific metric I mentioned. The issue is that we don’t restart the time counter over a snapshot restart, which panic when trying to use the pointer
Problem encountered
Validator panic'ed during restart, the stack trace is similar to #6373
Observed behaviour
n01.devnet1.vega.xys - is a validator server with vegavaisor. Every 30 min a random server is restarted, this single restart failed.
Expected behaviour
The server should restart.
Evidence
Logs
Additional context
Devnet 1 network:
Definition of Done
Before Merging
After Merging
Done
if there is NO requirement for new system-tests