Closed jgsbennett closed 10 months ago
The events that lead the the panic are:
If the rotation event had been found the system tests would call lnlMod.sync_genesis_with_network()
and all would be fine, but there is no recovery in any teardown that does this, so we wander into the next test with a bad network state.
So the panic is not the issue here. The unknown is why the key-rotation command didn't make it to core and this is difficult to know because 1) all the key rotation logs are set to debug 2) the network state is removed by the subsequent tests so I can't replay locally.
My suspicion is that its a timing issue? but I've got no evidence, so I'll just run the test in a loop a load to see if I can make it happen again.
Closing as this cant be reproduced
Problem encountered
We saw this panic during a test run. See slack thread: https://vegaprotocol.slack.com/archives/C03K403SZMM/p1693561610707909 It's really not clear what caused it, or if it's repeatable. There's weird things going on during this test run, so it's possibly caused by something devops-related. Raising in case somebody is able to deduce any more than me from the stack.
Observed behaviour
Panic
Expected behaviour
No Panics
Steps to reproduce
Software version
77e769f768e73c78df33c79df6f11332593aed53
Failing test
No response
Jenkins run
No response
Configuration used
No response
Relevant log output
No response