Sean, I think the freeze was not due to some recovery tests since they told
that they had to DROP the table to solve the problem. (ps: they do not mention
if they tried to do a gfix before dropping the table).

[]s

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

LS> Carlos,

LS> At the moment I'm inclined to think the "freeze" they reported was due
LS> to IB performing some recovery tasks, but can't say for sure. I've
LS> never tried this type of test or seen that "reaction" before.

LS> Perhaps, someone else could provide better/more informed feedback.


LS> Sean

LS> -----Original Message-----
LS> From: Carlos H. Cantu [mailto:warmbooter@...]
LS> Sent: Friday, January 26, 2001 11:14 AM
LS> To: Leyne, Sean
LS> Subject: Re[4]: [ib-support] Crash recovery test ! Keep
LS> reading...

LS> I will ask them why in the first test they tested both with FW ON and
LS> OFF
LS> and in the second test only with FW = OFF. Anyway, what do you think
LS> about
LS> the failured in the first test when FW was ON ?

LS> []s

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

LS>> Carlos,

LS>> I would strongly urge you to post a message to the editor pointing
LS> out
LS>> that the test was not a true representation of Interbase/Firebird
LS>> database integrity or recovery abilities.


LS>> Sean

LS>> -----Original Message-----
LS>> From: Carlos H. Cantu [mailto:warmbooter@...]
LS>> Sent: Friday, January 26, 2001 11:06 AM
LS>> To: Leyne, Sean
LS>> Subject: Re[2]: [ib-support] Crash recovery test ! Keep
LS>> reading...

LS>> Yes, I forgot to mention that FW was off in the 2nd test.

LS>> []s

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

LS>>> Carlos,

LS>>> In the second test, did they indicate that Forced Writes was ON?

LS>>> If not, than the fact that the database was "scrambled" would be an
LS>>> expected outcome.

LS>>> Database data integrity is only assured if Forced Writes = On.


LS>>> Sean

LS>>> -----Original Message-----
LS>>> From: Carlos H. Cantu [mailto:warmbooter@...]
LS>>> Sent: Friday, January 26, 2001 10:53 AM
LS>>> To: interbase@...
LS>>> Cc: ib-support@yahoogroups.com
LS>>> Subject: [ib-support] Crash recovery test ! Keep reading...

LS>>> Hello !

LS>>> I just found an article published in a Linux magazine in Brazil
LS>> about
LS>>> some
LS>>> tests with IB running on Linux related to IB crash recovery system.

LS>>> The article (in portuguese) can be found at
LS>>>
LS>>
LS> http://www.revistadolinux.com.br/artigos/database/20010117084829.php3

LS>>> Here is a resume of what they did. I would like to hear some
LS>> comments
LS>>> about
LS>>> their conclusions :

LS>>> They did basicly 2 kinds of crash tests :

LS>>> 1) Killing the main process (and its subprocess) using the linux
LS>> KILL
LS>>> command during some operations (insert,update,etc.)

LS>>> 2) RESETing the hardware during some operations
LS> (insert,update,etc.)

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

LS>>> In the second case, IB returned errors when trying to connect to
LS> the
LS>>> database :

LS>>> üdatabase file appear corrupt ( )
LS>>> wrong page type
LS>>> page "xxxxx" is of wrong type (expected 7, found 5)

LS>>> In some of them, the database could be connected but the table was
LS>>> inaccessible :

LS>>> Internal GDS software consistency check (cannot find record back
LS>> version
LS>>> (291)) (3 ocorrências)
LS>>> §Internal GDS software consistency check (wrong record length
LS> (183))
LS>>> §Internal GDS software consistency check (applied diferences will
LS>> not
LS>>> fit in record (177))
LS>>> §database file appear corrupt ( )
LS>>> wrong page type
LS>>> page "xxxxx" is of wrong type (expected 5, found 4)



LS>>> []s

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



LS>>> To unsubscribe from this group, send an email to:
LS>>> ib-support-unsubscribe@egroups.com



LS>>> To unsubscribe from this group, send an email to:
LS>>> ib-support-unsubscribe@egroups.com



LS>> To unsubscribe from this group, send an email to:
LS>> ib-support-unsubscribe@egroups.com



LS>> To unsubscribe from this group, send an email to:
LS>> ib-support-unsubscribe@egroups.com



LS> To unsubscribe from this group, send an email to:
LS> ib-support-unsubscribe@egroups.com



LS> To unsubscribe from this group, send an email to:
LS> ib-support-unsubscribe@egroups.com