Thanks Paul ! I'm trying to get the email of the authors so I can get more
details from them. I'll let you know.

[]s

Carlos
WarmBoot Informatica - http://www.warmboot.com.br
Interbase-BR - http://www.interbase-br.com

PR> "Leyne, Sean" wrote:

>>
>> In the first case, IB recovery system was 100% in almost all cases .
>> Only
>> in 1 situation an error was found : While updating 300,000 records with
>> forced write ON, after killing the process and trying to connect to the
>> database again, the connection was accepted and NO error message was
>> reported, but when the user tried to access the table that was being
>> updated the system freezed.
>>

PR> Did they say how long they waited for the 'system' to unfreeze? My first
PR> thought is that the attempt to open the 300,000 row table caused the engine to
PR> clean out back record versions that had not been committed. That can be quite
PR> lengthy on a table of that size.

PR> You/they didn't say whether the tests were against Classic or Superserver, or
PR> which version - 4, 5 or 6. Version 6 Super server ought not to have the
PR> 'cleanup' problem (at least, one of the SS features is that a separate thread
PR> will do this.) But v6 Classic will behave in the way described above.

PR> Also, system freezing can mean a multitude of things - from the interbase server
PR> process(es) to the kernel locking up, via the current terminal appearing to die.


PR> Paul