Subject | Re: [firebird-support] Some feedback of my 2.5 experience |
---|---|
Author | Lester Caine |
Post date | 2011-06-03T08:45:50Z |
Dunbar, Norman (Capgemini) wrote:
I think that while many people ae using FB.5 happily, there IS something that a
few of us are hitting? In my case I have systems that need to run reliably
during office hours and while I've moved a couple of sites over without a
problem, the last two upgrades had to be rolled back fast and there has been
nothing obvious in the logs. The only way to recreate the problem is to anoy the
sites with what are essentially random freezes during the day ... but not every
day. *IF* the traditional back step of backup up the 2.5 copy with 2.1 gbak had
not been lost, then it would be a lot easier to roll forward check things for a
couple of days and then roll back while investigating, but without that path I'm
loathed to move sites forward at the moment, so can't get the detail that is
needed to find the problem. What I can say though is that it certainly seems to
be related to the windows client software but the target server can be windows
or linux - fortunately or not the last two sites have different servers :)
--
Lester Caine - G8HFL
-----------------------------
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php
>>> If it is a deadlock, then it's a deadlock that's happeningIgnoring the 'demands' for more detail ...
>>> >> on FB 2.5, and NOT happening on FB 2.1
> I'd almost put money on it not being a deadlock. At least, not in the
> normal sense of the word as applied to the application. Maybe something
> internal is deadlocking (if that's at all possible?) - such as
> attempting to create/use a new database page for an index or data page
> or similar?
I think that while many people ae using FB.5 happily, there IS something that a
few of us are hitting? In my case I have systems that need to run reliably
during office hours and while I've moved a couple of sites over without a
problem, the last two upgrades had to be rolled back fast and there has been
nothing obvious in the logs. The only way to recreate the problem is to anoy the
sites with what are essentially random freezes during the day ... but not every
day. *IF* the traditional back step of backup up the 2.5 copy with 2.1 gbak had
not been lost, then it would be a lot easier to roll forward check things for a
couple of days and then roll back while investigating, but without that path I'm
loathed to move sites forward at the moment, so can't get the detail that is
needed to find the problem. What I can say though is that it certainly seems to
be related to the windows client software but the target server can be windows
or linux - fortunately or not the last two sites have different servers :)
--
Lester Caine - G8HFL
-----------------------------
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk//
Firebird - http://www.firebirdsql.org/index.php