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

Re: failed commit not totaly rolled back

From: Carl Karsten <carl_at_personnelware.com>
Date: 2005-10-18 20:37:50 CEST

Joseph Galbraith wrote:
> Philip Martin wrote:
>
>> Carl Karsten <carl@personnelware.com> writes:
>>
>>> Adding doc
>>> Adding (bin) doc\bp.odg
>>> Transmitting file data ..svn: Commit failed (details follow):
>>> svn: Error processing command 'committed' in 'C:\vfe\dev\WinPars\DATA'
>>> svn: Error replacing text-base of 'ErrorLog.FPT'
>>> svn: Can't copy
>>> 'C:\vfe\dev\WinPars\DATA\.svn\tmp\text-base\ErrorLog.FPT.svn-base' to
>>> 'C:\vfe\dev\WinPars\DATA\.svn\tmp\ErrorLog.FPT.tmp.tmp': There is not
>>> enough space on the disk.
>>>
>>> So I dump the 130mb of /temp files and try again.
>>
>>
>>> Adding doc
>>> Adding (bin) doc\bp.odg
>>> Transmitting file data ..svn: Commit failed (details follow):
>>> svn: Out of date: 'DATA/ErrorLog.FPT' in transaction '16-1'
>>>
>>> It seems when the first attemt failed it didn't completely roll back
>>> everything it had started, which I am guessing is a bug?
>>
>>
>> No. The first commit failed during post-commit processing of the
>> working copy, that's after the commit completed in the repository.
>> The failure means the working copy is out-of-date compared to the
>> repository, so the second commit fails.
>
>
> Definitely an error message that is a little confusing though...
> maybe it could be tweaked a bit to say that the commit succeeded,
> but the working copy could not be updated?
>

Why is there "post-commit processing" ? shouldn't all processing be part of the
transaction such that if any parts fail, the whole thing gets rolled back?

^Carl

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Oct 18 20:38:38 2005

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

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