Hans-Emil Skogh <Hans-Emil.Skogh_at_tritech.se> wrote:
> >> From your description I guess this is an issue with a deadlock
> >> somewhere in the cache. I have a look at this, but deadlocks are hard
to
> >> find and to reproduce so this might take me a while to fix...
> > I got it once more today, and when I tried a little (aggressively
browsing
> > in explorer) I managed to provoke it a second time.
> > When the nightlies are built I'll download the debug symbols and see
if
> > I can attach to the TSVNCache process the next time it stalls.
>
> I got it again a couple of times yesterday. I successfully attached the
debugger to TSVNCache.exe,
> but didn't get that much wiser...
>
> Stefan^2, any hints on where to look? Any suspicions?
For a couple for days now, I had a similar issue
with my XP notebook but no time to look into it.
I just happened to encounter a slightly different
issue on my Win7 office PC: The TSVNCache process
would occupy one core and the OutputDebugString
log would show timeouts.
The reason was / is that I had 1.7dev working copies
in some older format and the SVN libs silently upgrade
them to format 24. That obviously takes quite some
time (minutes to hours per w/c). After that, everything
seems to be back to normal.
Tonight, I will check whether it's the same issue
on my notebook.
-- Stefan^2.
------------------------------------------------------
http://tortoisesvn.tigris.org/ds/viewMessage.do?dsForumId=757&dsMessageId=2706150
To unsubscribe from this discussion, e-mail: [dev-unsubscribe_at_tortoisesvn.tigris.org].
Received on 2011-02-21 15:25:11 CET