Comment 80 for bug 1890435

r10-6080 now had 10 good runs.

I'm going back to test 20200507 next - we had bad states with that version so often this MUST trigger IMHO,

Reminder this runs on in armhf LXD containers on arm64 VMs (like our builds do).
I'm slowly getting the feeling it could be an issue with the underlying virtualization or bare metal.
We had a datacenter move, so the cloud runs on the same bare metal overall, but my instance could run on something else today than last week. If 20200507 no more triggers we have to investigate where the code is running.

20190425 good
r10-1014
r10-2027 good
r10-2533
r10-3040 good
r10-3220
r10-3400 good
r10-3450
r10-3475
r10-3478
r10-3593
r10-3622
r10-3657 good
r10-3727 good
r10-4054 other kind of bad?
r10-6080 good
20200507 bad ?next?