On Fri, 14 Apr 2023, Laurenz Albe wrote:
>So if your disk replaces a valid block with zeros (filesystem check
>after crash?), that could explain what you see.
Oh, I had that happen on a RAID 1 once. On of the two discs had an
intermittent error (write I guess) but didn’t fail out of the RAID,
and some of the reads from there got zero-filled blocks in some
positions. It was a CVS repository so I was able to identify all such
blocks in question and restore them from the rsync slave (whose initial
population predated the HDD issue).
Hm, now that I think about it, it could even have been a read error
with subsequent block reassignment. Oh well.
Filesystem issues (ext3, and ext4 without/predating auto_da_alloc,
in particular) could be it just as well of course.
bye,
//mirabilos
--
15:41⎜<Lo-lan-do:#fusionforge> Somebody write a testsuite for helloworld :-)