Please describe what you are doing to trigger the bug:
- I ran
check -chunks
on a storage and it tells me I have a corrupted chunk. - To check the rest of the chunks I re-ran it with
check -chunks -persist
. This reports I have 7 corrupted chunks. However it also tells me it added all chunks as verified to the “known good” list. - I now again run
check -chunks
and it reports no corrupted chunks.
Please describe what you expect to happen (but doesn’t):
I would expect check -chunks -persist
to check all chunks and report the corrupted ones. I would then expect the good ones to be added to the “known good” list. I would expect the next run of check
to still report the corrupted chunks as corrupted.
Please describe what actually happens (the wrong behaviour):
The corrupted chunks are added to the “known good” list even if they failed verification. Subsequent runs of check now incorrectly states everything is OK when it is not.