So far everything works as it should and I did not manage to observe this problem again. I just rejoined that node (resynced data) to get it back operational.
Regarding file corruption - what I am pretty sure about is it could not be hardware corruption. The OS does not report any filesystem errors, the server runs two brand new Intel SSDs in Raid1.
So far everything works as it should and I did not manage to observe this problem again. I just rejoined that node (resynced data) to get it back operational.
Regarding file corruption - what I am pretty sure about is it could not be hardware corruption. The OS does not report any filesystem errors, the server runs two brand new Intel SSDs in Raid1.