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

Re: rfc for proposal on solving issue #443

From: Marcus Rueckert <darix_at_web.de>
Date: 2005-05-05 13:58:54 CEST

On 2005-05-05 04:45:28 -0700, Madan US wrote:
> > > #443 is about propagating the post-commit hook failure back to the
> > does the client really wait until the post-commit is ran? i thought it
> > would disconnect before the hook is ran.
>
> I dont think it happens that way.
> Inside dav_svn_merge(), The svn_repos_fs_commit_txn() function ( which
> internally exectues the post-commit hook ) is called before calling
> dav_svn__merge_response() which creates the merge-response.
> And the run_hook_cmd() waits for the post-commit hook to return.
 
strange... is this documented for hook authors somewhere?
i dont see it in the book. this is bad. because the book mentions
hotbackup.py ... and this script can use quite some time for larger
repositories.
> > i would think it can cause a bit trouble if you have long running
> > post-commits.
>
> true... delays may also cause timeouts!!!
thats what i thought.
> > i think simple logging of those might be better, but that
> > is stuff for the hook script authors.
>
> True, the hook script authors have to take care of this.
> btw, what do you think of the idea I mentioned?!
in general... a normal user dont need to know about such scripts and
what they returned. it is a admin side only thing. so if an admin wants
to track his script he should add logging to it.
darix

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu May 5 14:00:13 2005

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