Our build server isn’t feeling great; it has a failed disk and another one that’s acting suspiciously and, although there is redundancy, apparently some data was eaten. The Windows VM specifically is down:
2022-10-28T10:05:14.821637Z qemu-system-x86_64: -drive file=/data/vm/win10.img,format=qcow2,if=none,id=drive-virtio-disk0,cache=none,aio=native:
qcow2: Image is corrupt; cannot be opened read/write
Our non-Windows builds run on a different RAID volume which isn’t affected. I’ll think on what to do about it. Replacing the failed disk(s) is of course an option but it’s in another town and juggling hardware isn’t my idea of fun. More cloud may be added to the mix instead…