You're right. Looking at the 1.4.14 source and comparing with what the
patches intended, it's clear the patch(es) never made it into the 1.4.14
release.
There are patches here
http://web.mit.edu/jaltman/Public/OpenAFS/patches/vlserver-volid-1.patch
and here
http://rt.central.org/rt/index.html?q=124510
I've applied the web.mit.edu set of patches, and I've made a patch to to
src/vol/vol-salvage.c which is required for the salvager to work, but
may have been included in the second set (rt.central.org)
I'll revert to the distributed 1.4.14 source and apply both sets of
patches above, and report back.
Kim
On 3/9/2011 3:18 PM, Andrew Deason wrote:
On Wed, 09 Mar 2011 14:49:24 -0700
Kim Kimball<[email protected]> wrote:
From what I could gather the "invalid volume id" bug was first a patch
and then rolled in to 1.4.14
The only changes in 1.4.14 that were not in 1.4.12.1 were security fixes
and a couple of small platform-specific things, so I don't think so. I
also don't recall seeing any patches for anything like this recently.
That said, that's quite a large volume id (greater than 2^31). I
wouldn't be surprised if you find something else that has an issue with
that besides the salvager.
03/09/2011 13:42:41 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager
/vicepa 3249855668)
03/09/2011 13:42:41 salvage: invalid volume id specified; salvage aborted
Ugh, this check is awful. Assuming we want to keep it, it seems like it
should be an isdigit loop or something instead.
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info
