[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svnadmin verify changing its diagnosis?

From: John Szakmeister <john_at_szakmeister.net>
Date: 2006-10-07 02:19:57 CEST

----- Rob Hubbard <Rob.Hubbard@celoxica.com> wrote:
> I'm afraid that this is not an answer to your question.
>
> Are you using an filing system or a database backend? (I believe that
> this is indicated by the contents of the file db/fs-type in the
> repository.)
>
> Also, did you see the problem more than once?
>
> The reason for asking is that, if I understand correctly, the DB
> backend stores changes as reverse deltas backwards from the head
> revision, whereas the FS backend stores them as forwards deltas
> forwards from revision 0. So, only the database backend recalculates a
> chain of deltas during a commit.
>
> I wonder if it's possible that the problem has disappeared due to a
> faulty delta being recalculated (following a commit).
>
> Alternatively, I wonder whether it's possible to run svnadmin verify
> during a commit, and that what you saw was a repository database in a
> transient state.
>
> Does anyone know?

svnadmin verify records what the latest committed revision is, and verifies from 0 to that revision. The revisions are immutable, and are transaction based, so there should be no issues running svnadmin verify during a commit operation.

-John

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Sat Oct 7 02:20:36 2006

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.