On 16/07/10 11:38, Stefan Sperling wrote: > On Fri, Jul 16, 2010 at 11:29:46AM +0100, John Beranek wrote: >> On 16/07/10 09:45, Stefan Sperling wrote: >>> On Thu, Jul 15, 2010 at 11:47:48PM +0100, John Beranek wrote: >>>> >> [snip] >>>> Has no-one else reported this? Is it being worked on - anything? >>> >>> Looks like this was improved in trunk ages ago, but not backported >>> to 1.6.x. Can you try patching a client with the diff below and let >>> us know if this fixes the problem? >> >> Added the patch to the Fedora 12 subversion-1.6.11 SRPM and built it. It >> does indeed fix the error reporting. >> >> Maybe I can try and get this patch applied in Fedora distros for a start... > > I've nominated the patch for backport to 1.6.13: > http://svn.apache.org/viewvc/subversion/branches/1.6.x/STATUS?r1=964473&r2=964736
Out of interest, does ra_serf behave "nicely" too. A "grep 403 *" in both the trunk code and the 1.6.11 code I have here suggest the answer may be "no". 403 is mentioned in only commit.c and locks.c, so perhaps write failures will give appropriate warnings, but not read failures? I haven't got an ra_serf enabled build of Subversion here to test. (Fedora (12) doesn't have a serf package) Cheers, John. -- John Beranek To generalise is to be an idiot. http://redux.org.uk/ -- William Blake
smime.p7s
Description: S/MIME Cryptographic Signature
