> > You can try this snapshot of 2.1.4 with patch for index corruptions :
> >
> > http://www.firebirdsql.org/download/rabbits/hvlad/Firebird-2.1.4.18274_Win32.zip
>
> I will try ! but is it the 64bit version? can i try it in our production server ?

x64 build is there :

http://www.firebirdsql.org/download/rabbits/hvlad/Firebird-2.1.4.18274_x64.zip

This is snapshot build based on current state B2_1_Release branch, i.e. it is 2.1.3 plus all committed bugfixes plus new patch for index corruption issue. It is not tested by our QA team as such tests usually runs before making public release. I know that at least 3 customers now tested it in production and 1 of them already reported that indices are ok with this build.

> > Follow instructions at
> >
> > http://www.ibphoenix.com/main.nfs?a=ibphoenix&s=1254244067:355713&l=;PAGES;NAME=%27ibp_pdb_win32%27
> >
> > to produce crash dump for us.
>
> ok, i will do... do you know where i can found the pdb (Debug) Firebird installation ?

You can found .pdb files at the same location where you download official release. For example for v2.1.3 download page

http://www.firebirdsql.org/index.php?op=files&id=engine_213

contains so called "Debug Kits (Binary + PDB components)" for both Win32 and Win64 platforms.

Note, .pdb files should correspond to the actual .exe files. My build (above) already contains necessary .pdb files.

> > > It look like also than when we reduce the number of page buffer then the crash is not so often (for exemple with a page buffer of 75)
> > >
> > > as i know, every page need to be "lock", so more page buffer mean more lock and in classic server they are multiply by the number of client, so around 2048 page buffer * 500 connections = 1 000 000 page to lock... is it possible that it's was a lock problem ??
> >
> > Output of fb_lock_print -h could help to answer
>
> below the result of the fb_lock_print when the problem was (i don't remembered if i launch it with the -h)

I see no issues with lock manager

Regards,
Vlad