On 11/08/2014 06:05 PM, Andrew Deason wrote:
That particular warning doesn't seem like it would cause problems; just
taht we're using up extra unused space in the db. Would you be willing
to share your (partially-corrupt) vldb blob? That way we could try to
make vldb_check handle that issue.
Oh, good. I though so but it is a relief to read it from somebody who
knows better. My database is not big, about 1.5MB is all, but if this
may help other cells which far bigger databases I am all in for helping
you guys resolve it.
I just mailed my vldb to Mike to test his patch. I can forward you a
copy if you like as well.
It does seem like maybe a better way to fix corruption errors would be
to dump the vldb to a text format, and be able to restore that back to a
db blob. But, that's not what we have now.
Right, this is what I had in mind. I had done the same in the past,
prompted by people in this list, with pt_util for my protection database
to remove some entries prdb_check -fix woundn't touch. So when I saw the
dump option in there it rung a bell.
Thank you all for your time.
-Kostas
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info