Hi Sergei,
On Mon, Aug 13, 2007 at 07:40:52PM +0400, Sergei Vyshenski wrote:
> Please note that tests pass with LESS sleeping time and fail
> with MORE sleeping time. Do you think that less time
> could model faster termination of the Dumper operation called
> from this line?
> All this could explain why tests pass ok at your testing machine
> because yours in somewhat faster than ours.
Yes, that might be a reason. It could be a timeout on the server
which causes the client to hang or so ...
> Could you please try to reproduce our problem on your host
> placing "sleep 20" or better "sleep 60" to this place
> instead of the line 159?
I'll try that soon, Martin and I are still a bit busy with a production
deployment here.
> Looks like OpenXPKI's Persister module has problems
> with handling of the time. We have just tried today's patch
> of Jim Brandt for persister part of the Workflow module, and
> this patch does not change anything related to our picture.
> Nonetheless this patch looks very much like really related to
> our problem, but we do not understand how use this knowledge
> to fix the problem.
Does it? I believe it deals with date/time formats, so I don't
really see how this is related to what we do ...? We even do the
time handling ourself in our DBI persister.
> 2. Another very interesting point. This is about output of
> several tests.
> For example consider a test t/25_crypto/11_use_ca.
>
> In the output (file 1034_output) of your (Linux) testing host I can see
>
> [...]
>
> Please make notice how the ORDER of exception reports and of OK statements
> are different between Linux and FreeBSD.
> Looks like FreeBSD follows the sequence of the perl source,
> while Linux violates this sequence, maybe because of some parallel
> processing or so.
>
> How have your achieved this?
> Do you think that this parallelism could be responsible for time races
> or such?
No, I don't. I believe the diag statements (the ones starting with '#')
are output on STDERR and the rest is output on STDOUT. This looks like
some autoflush issue which causes the filehandles to be flushed at
different times. Nothing to worry about, IMHO.
Best regards,
Alex
--
Dipl.-Math. Alexander Klink | IT-Security Engineer
[EMAIL PROTECTED] | working @ urn:oid:1.3.6.1.4.1.11417
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
OpenXPKI-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openxpki-devel
