Some guess or hint

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Some guess or hint

Jiří Činčura-2
Hi *,

I'm almost closing on compression support. It is stable when I run the
tests on my machine. Multiple time, 32bit, 64bit. But are failing on
AppVeyor.

To make matters worse it's always different test(s). Check
https://ci.appveyor.com/project/cincura_net/firebirdsql-data-firebirdclient/build/880
or
https://ci.appveyor.com/project/cincura_net/firebirdsql-data-firebirdclient/build/882
. Only common piece is that it's always test with compression on (2nd
parameter is True).

I spent few days trying to guess what's wrong and fixing some things.
But no final luck (I'll not list mine to not skew the thinking). So
maybe some smart people here will have some idea.

So let's open discussion. Or if you can make it fail on you machine,
that would help as well.

--
Mgr. Jiří Činčura
Independent IT Specialist

------------------------------------------------------------------------------
_______________________________________________
Firebird-net-provider mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/firebird-net-provider
zab
Reply | Threaded
Open this post in threaded view
|

Re: Some guess or hint

zab
I had catched such random issues by couple of tricks:
1. .trx files that were produced vstest.console. They always have call
stack with exception.
2. Unhandled exception handler that creates dump file of process
itself and then publish it as artifact. Dump can be opened in Visual
Studio, it will look as attached to process. Though I prefer using of
WinDbg

On 27 August 2016 at 10:28, Jiří Činčura <[hidden email]> wrote:

> Hi *,
>
> I'm almost closing on compression support. It is stable when I run the
> tests on my machine. Multiple time, 32bit, 64bit. But are failing on
> AppVeyor.
>
> To make matters worse it's always different test(s). Check
> https://ci.appveyor.com/project/cincura_net/firebirdsql-data-firebirdclient/build/880
> or
> https://ci.appveyor.com/project/cincura_net/firebirdsql-data-firebirdclient/build/882
> . Only common piece is that it's always test with compression on (2nd
> parameter is True).
>
> I spent few days trying to guess what's wrong and fixing some things.
> But no final luck (I'll not list mine to not skew the thinking). So
> maybe some smart people here will have some idea.
>
> So let's open discussion. Or if you can make it fail on you machine,
> that would help as well.
>
> --
> Mgr. Jiří Činčura
> Independent IT Specialist
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Firebird-net-provider mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/firebird-net-provider

------------------------------------------------------------------------------
_______________________________________________
Firebird-net-provider mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/firebird-net-provider
Reply | Threaded
Open this post in threaded view
|

Re: Some guess or hint

Jiří Činčura-2
Too late. It's already solved. It was a race condition in Firebird
itself.

--
Mgr. Jiří Činčura
Independent IT Specialist

------------------------------------------------------------------------------
_______________________________________________
Firebird-net-provider mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/firebird-net-provider