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

Re: DB_RUNRECOVERY: Fatal error, run database recovery

From: Stefan Sperling <stsp_at_elego.de>
Date: Thu, 16 Dec 2010 19:46:40 +0100

On Sat, Dec 11, 2010 at 10:00:29PM +0000, HM 2K wrote:
> I'm having some trouble accessing the svn via Apache, with the following
> error:
> "Could not open the requested SVN filesystem"
> I've recently upgraded CentOS from 4 to 5 and upgraded SVN along with it.
> I have verified and recovered the database, both report back OK now.
> Can you advise further?
> Thanks.

See http://subversion.apache.org/faq.html#stuck-bdb-repos

You could convert your repository to FSFS to avoid this problem.
This is done by telling everyone to stop committing, and then
dumping the repository:
  svnadmin dump oldrepos > oldrepos.dump
... creating a new repository with FSFS backend (FSFS is default
in current releases of Subversion):
  svnadmin create --fs-type=fsfs mynewrepository
... and then loading the dump file into the new repository:
  svnadmin load mynewrepository < oldrepos.dump
... and replacing the old repository with the new one.

You may also want to set the new repository's UUID to the UUID of the
old one:
  svnadmin setuuid mynewrepository `cat oldrepos/db/uuid`
This will keep existing working copies working.

Received on 2010-12-16 19:47:31 CET

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.