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...@shadyvale.net> on 2003/08/16 16:44:45 UTC

Tests on Windows

A number of ra_local tests are failing on Windows, even on two machines 
that did not fail any tests a week ago or so.  It could be the 
filesystem problem showing up, of course, but last run (week or so ago) 
I had no failures at all on two machines.

Could someone else run tests on Windows and confirm/deny the results? 
All three machines fail the same five tests consistently after 3 runs 
each (same binaries).  The full logs are posted at 
http://www.shadyvale.net/svn/.

DJ


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

Re: Tests on Windows

Posted by "D.J. Heap" <dj...@shadyvale.net>.
cmpilato@collab.net wrote:
> cmpilato@collab.net writes:
> 
> 
>>"D.J. Heap" <dj...@shadyvale.net> writes:
>>
>>
>>>A number of ra_local tests are failing on Windows, even on two
>>>machines that did not fail any tests a week ago or so.  It could be
>>>the filesystem problem showing up, of course, but last run (week or so
>>>ago) I had no failures at all on two machines.
>>
>>I just updated my Win2k box to HEAD, and the five tests your results
>>show failing had these results for me:
>>
>>   commit_tests.py 22 : PASS
>>   commit_tests.py 23 : PASS
>>   update_tests.py 15 : FAIL  (no error message, just expected results)
>>   merge_tests.py   2 : FAIL  (access denied opening 'tmp.2')
>>   merge_tests.py   8 : FAIL  (access denied opening 'tmp.2')
> 
> 
> Oh, do note that I did *not* pass the -r so that test data was created
> outside the source tree.  That might be a clue as to the two
> commit-tests failures.
> 
> Also (to my relief), I get exactly the same results when using
> Subversion revision 6751 (just prior to the FS schema changes merge).


Yes, sorry, I've been out of town, but I did not mean I thought the FS 
scheme changes were at fault...I was referring to the Windows filesystem 
problems we've been seeing and was mostly just asking if someone else 
would/has been running tests and what results they were getting to 
compare with mine.

I'm trying to run them at least once a week and when I get more failures 
than I can be fairly certain are symptoms of the known filesystem 
problem, I try to email the list to see if I'm alone in the failures or 
what.

Thanks for the results!

DJ


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

Re: Tests on Windows

Posted by cm...@collab.net.
cmpilato@collab.net writes:

> "D.J. Heap" <dj...@shadyvale.net> writes:
> 
> > A number of ra_local tests are failing on Windows, even on two
> > machines that did not fail any tests a week ago or so.  It could be
> > the filesystem problem showing up, of course, but last run (week or so
> > ago) I had no failures at all on two machines.
> 
> I just updated my Win2k box to HEAD, and the five tests your results
> show failing had these results for me:
> 
>    commit_tests.py 22 : PASS
>    commit_tests.py 23 : PASS
>    update_tests.py 15 : FAIL  (no error message, just expected results)
>    merge_tests.py   2 : FAIL  (access denied opening 'tmp.2')
>    merge_tests.py   8 : FAIL  (access denied opening 'tmp.2')

Oh, do note that I did *not* pass the -r so that test data was created
outside the source tree.  That might be a clue as to the two
commit-tests failures.

Also (to my relief), I get exactly the same results when using
Subversion revision 6751 (just prior to the FS schema changes merge).

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

Re: Tests on Windows

Posted by "D.J. Heap" <dj...@shadyvale.net>.
Commit 22 is one that seems to be a symptom of the windows filesystem 
problem, it fails sporadically, but more often on my faster machines, 
and rarely on the slower ones (I seem to recall thinking that the chdir 
failing would cause it?)...That one fails for me often (along with 23, 
though that one is fixed now, hopefully).  It was the 3 new failures 
that worried me.  I'll update and rerun tests tonight.

Thanks!

DJ


Philip Martin wrote:

> cmpilato@collab.net writes:
> 
> 
>>"D.J. Heap" <dj...@shadyvale.net> writes:
>>
>>
>>>A number of ra_local tests are failing on Windows, even on two
>>>machines that did not fail any tests a week ago or so.  It could be
>>>the filesystem problem showing up, of course, but last run (week or so
>>>ago) I had no failures at all on two machines.
>>
>>I just updated my Win2k box to HEAD, and the five tests your results
>>show failing had these results for me:
>>
>>   commit_tests.py 22 : PASS
> 
> 
> I don't know why test 22 fails for D.J., but it's a commit that uses
> an absolute path.  Could it be something to do with the Windows drive
> letters?
> 
> 
>>   commit_tests.py 23 : PASS
> 
> 
> When test 22 fails it didn't reset the current working directory,
> causing test 23 to fail and none of the subsequent tests to run.
> Revision 6758 should fix that.
> 
> 
>>   update_tests.py 15 : FAIL  (no error message, just expected results)
>>   merge_tests.py   2 : FAIL  (access denied opening 'tmp.2')
>>   merge_tests.py   8 : FAIL  (access denied opening 'tmp.2')
> 
> 



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

Re: Tests on Windows

Posted by Philip Martin <ph...@codematters.co.uk>.
cmpilato@collab.net writes:

> "D.J. Heap" <dj...@shadyvale.net> writes:
>
>> A number of ra_local tests are failing on Windows, even on two
>> machines that did not fail any tests a week ago or so.  It could be
>> the filesystem problem showing up, of course, but last run (week or so
>> ago) I had no failures at all on two machines.
>
> I just updated my Win2k box to HEAD, and the five tests your results
> show failing had these results for me:
>
>    commit_tests.py 22 : PASS

I don't know why test 22 fails for D.J., but it's a commit that uses
an absolute path.  Could it be something to do with the Windows drive
letters?

>    commit_tests.py 23 : PASS

When test 22 fails it didn't reset the current working directory,
causing test 23 to fail and none of the subsequent tests to run.
Revision 6758 should fix that.

>    update_tests.py 15 : FAIL  (no error message, just expected results)
>    merge_tests.py   2 : FAIL  (access denied opening 'tmp.2')
>    merge_tests.py   8 : FAIL  (access denied opening 'tmp.2')

-- 
Philip Martin

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

Re: Tests on Windows

Posted by cm...@collab.net.
"D.J. Heap" <dj...@shadyvale.net> writes:

> A number of ra_local tests are failing on Windows, even on two
> machines that did not fail any tests a week ago or so.  It could be
> the filesystem problem showing up, of course, but last run (week or so
> ago) I had no failures at all on two machines.

I just updated my Win2k box to HEAD, and the five tests your results
show failing had these results for me:

   commit_tests.py 22 : PASS
   commit_tests.py 23 : PASS
   update_tests.py 15 : FAIL  (no error message, just expected results)
   merge_tests.py   2 : FAIL  (access denied opening 'tmp.2')
   merge_tests.py   8 : FAIL  (access denied opening 'tmp.2')

On Linux:

   commit_tests.py 22 : PASS
   commit_tests.py 23 : PASS
   update_tests.py 15 : PASS
   merge_tests.py   2 : PASS
   merge_tests.py   8 : PASS

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

Re: Tests on Windows

Posted by "D.J. Heap" <dj...@shadyvale.net>.
Revision 6753 -- fs changes plus some cleanup.

DJ


Erik Hülsmann wrote:
> 
> Is that before or after revision 6752 fs schema change?
> 
> bye,
> 
> 
> Erik.


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

Re: Tests on Windows

Posted by Erik Hülsmann <e....@gmx.net>.

Is that before or after revision 6752 fs schema change?

bye,


Erik.

>A number of ra_local tests are failing on Windows, even on two machines 
>that did not fail any tests a week ago or so.  It could be the 
>filesystem problem showing up, of course, but last run (week or so ago) 
>I had no failures at all on two machines.
>
>Could someone else run tests on Windows and confirm/deny the results? 
>All three machines fail the same five tests consistently after 3 runs 
>each (same binaries).  The full logs are posted at 
>http://www.shadyvale.net/svn/.
>
>DJ
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
>For additional commands, e-mail: dev-help@subversion.tigris.org

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