disk crashes provoked by tahoe-lafs?

Greg Troxel gdt at ir.bbn.com
Tue Apr 15 15:13:24 UTC 2014

Paul Rabahy <prabahy at gmail.com> writes:

> I don't care what you do, if a program operating in userspace can kill a
> disk, I would call that a hardware problem. Even if the hardware didn't
> want to claim the problem then the filesystem should be responsible for it.

I see your point, and I think it's 90% valid but a bit extreme.  A
user-space process in a tight loop doing create/fsync/unlink or
something like that could cause trouble.   There are lots of other
systems running similar software but without tahoe, and I'm not seeing
nearly as many failures on them - hence my query.

> My honest guess is that you were just hit with a string of bad luck. Were
> the drives that failed the same make/model/manufacture date?

No, not at all the same.  One was a Seagate 1T, ST1000DM03.  One was a
2T WD Elements (in a box, new in 2010).  One was a hitachi 500G notebook
drive HTS545050B9A300 (2.5", mounted in a net5501).  And one a Seagate
750G ST3750528AS.

The bad luck theory is quite plausible, and I put considerable weight on
it.  I was really asking if anyone else had experiences that
corroborated the theory.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 180 bytes
Desc: not available
URL: <http://tahoe-lafs.org/pipermail/tahoe-dev/attachments/20140415/49084a77/attachment.asc>

More information about the tahoe-dev mailing list