You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by "D.J. Heap" <dj...@gmail.com> on 2006/11/24 16:43:47 UTC

ra_dav test failures and crashes

Can anyone else confirm or deny that they are seeing lots of test
failures and crashes over ra_dav (r22416) on trunk?  ra_local works
fine.

I'll dig around more this weekend when I can, but would like
confirmation that it's not just my build or something...

DJ

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: ra_dav test failures and crashes

Posted by "D.J. Heap" <dj...@gmail.com>.
On 11/24/06, Erik Huelsmann <eh...@gmail.com> wrote:
[snip]
> Well, r22424 did solve a problem, but not yours (as discussed in IRC).
>
> r22430 however should have fixed it: it turned out to be a
> pool-destruction ordering problem caused by the fact that the
> session-pool was not a parent of the pool the request was allocated
> in.


Thanks for the fixes Erik, they are getting much further now.  The
tests all finished but I am still getting a few failures -- I don't
know if they are related or not, though.

I've attached the detailed log in case anyone wants to look at it
closer.  Hopefully I'll have more time this weekend to look at them as
well.

Thanks!

DJ

Re: ra_dav test failures and crashes

Posted by Erik Huelsmann <eh...@gmail.com>.
On 11/24/06, Erik Huelsmann <eh...@gmail.com> wrote:
> On 11/24/06, Erik Huelsmann <eh...@gmail.com> wrote:
> > On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> > > On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> > > > Can anyone else confirm or deny that they are seeing lots of test
> > > > failures and crashes over ra_dav (r22416) on trunk?  ra_local works
> > > > fine.
> > > >
> > > > I'll dig around more this weekend when I can, but would like
> > > > confirmation that it's not just my build or something...
> > > >
> > >
> > >
> > > Here's a stack trace:
> > >
> > > >       svn.exe!remove_hook
> > >         svn.exe!ne_unhook_pre_send
> > >         svn.exe!ne_decompress_destroy
> > >         svn.exe!compressed_body_reader_cleanup
> > >         libapr.dll!run_cleanups
> > >         libapr.dll!apr_pool_destroy
> > >         libapr.dll!apr_pool_destroy
> > >         svn.exe!svn_cl__checkout
> > >         svn.exe!main
> > >         svn.exe!__tmainCRTStartup
> > >         svn.exe!mainCRTStartup
> > >         kernel32.dll!_BaseProcessStart@4
> > >
> > > The hooks parameter passed to remove_hook is pointing to already-freed
> > > memory, so the while loop is crashing immediately.
> > >
> > > Also, I went back to r22345 from a few days ago and the tests are
> > > running clean so it's not just my build.  I'll dig deeper when I can.
> >
> > I'm affraid I'm to blame.
> > I'll fix.
>
> Should be fixed with r22424.

Well, r22424 did solve a problem, but not yours (as discussed in IRC).

r22430 however should have fixed it: it turned out to be a
pool-destruction ordering problem caused by the fact that the
session-pool was not a parent of the pool the request was allocated
in.

bye,

Erik.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: ra_dav test failures and crashes

Posted by Erik Huelsmann <eh...@gmail.com>.
On 11/24/06, Erik Huelsmann <eh...@gmail.com> wrote:
> On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> > On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> > > Can anyone else confirm or deny that they are seeing lots of test
> > > failures and crashes over ra_dav (r22416) on trunk?  ra_local works
> > > fine.
> > >
> > > I'll dig around more this weekend when I can, but would like
> > > confirmation that it's not just my build or something...
> > >
> >
> >
> > Here's a stack trace:
> >
> > >       svn.exe!remove_hook
> >         svn.exe!ne_unhook_pre_send
> >         svn.exe!ne_decompress_destroy
> >         svn.exe!compressed_body_reader_cleanup
> >         libapr.dll!run_cleanups
> >         libapr.dll!apr_pool_destroy
> >         libapr.dll!apr_pool_destroy
> >         svn.exe!svn_cl__checkout
> >         svn.exe!main
> >         svn.exe!__tmainCRTStartup
> >         svn.exe!mainCRTStartup
> >         kernel32.dll!_BaseProcessStart@4
> >
> > The hooks parameter passed to remove_hook is pointing to already-freed
> > memory, so the while loop is crashing immediately.
> >
> > Also, I went back to r22345 from a few days ago and the tests are
> > running clean so it's not just my build.  I'll dig deeper when I can.
>
> I'm affraid I'm to blame.
> I'll fix.

Should be fixed with r22424.

bye,

Erik.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: ra_dav test failures and crashes

Posted by Erik Huelsmann <eh...@gmail.com>.
On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> > Can anyone else confirm or deny that they are seeing lots of test
> > failures and crashes over ra_dav (r22416) on trunk?  ra_local works
> > fine.
> >
> > I'll dig around more this weekend when I can, but would like
> > confirmation that it's not just my build or something...
> >
>
>
> Here's a stack trace:
>
> >       svn.exe!remove_hook
>         svn.exe!ne_unhook_pre_send
>         svn.exe!ne_decompress_destroy
>         svn.exe!compressed_body_reader_cleanup
>         libapr.dll!run_cleanups
>         libapr.dll!apr_pool_destroy
>         libapr.dll!apr_pool_destroy
>         svn.exe!svn_cl__checkout
>         svn.exe!main
>         svn.exe!__tmainCRTStartup
>         svn.exe!mainCRTStartup
>         kernel32.dll!_BaseProcessStart@4
>
> The hooks parameter passed to remove_hook is pointing to already-freed
> memory, so the while loop is crashing immediately.
>
> Also, I went back to r22345 from a few days ago and the tests are
> running clean so it's not just my build.  I'll dig deeper when I can.

I'm affraid I'm to blame.
I'll fix.

bye,

Erik.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: ra_dav test failures and crashes

Posted by "D.J. Heap" <dj...@gmail.com>.
On 11/24/06, D.J. Heap <dj...@gmail.com> wrote:
> Can anyone else confirm or deny that they are seeing lots of test
> failures and crashes over ra_dav (r22416) on trunk?  ra_local works
> fine.
>
> I'll dig around more this weekend when I can, but would like
> confirmation that it's not just my build or something...
>


Here's a stack trace:

>	svn.exe!remove_hook
 	svn.exe!ne_unhook_pre_send
 	svn.exe!ne_decompress_destroy
 	svn.exe!compressed_body_reader_cleanup
 	libapr.dll!run_cleanups
 	libapr.dll!apr_pool_destroy
 	libapr.dll!apr_pool_destroy
 	svn.exe!svn_cl__checkout
 	svn.exe!main
 	svn.exe!__tmainCRTStartup
 	svn.exe!mainCRTStartup
 	kernel32.dll!_BaseProcessStart@4

The hooks parameter passed to remove_hook is pointing to already-freed
memory, so the while loop is crashing immediately.

Also, I went back to r22345 from a few days ago and the tests are
running clean so it's not just my build.  I'll dig deeper when I can.

DJ

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org