So far I haven't found any operation that could have resulted in an
endless loop.

But I found an interesting clue about the error state: when it
happened, calling a simple select stored procedure in that database
will return the error, but calling that same procedure for a copy of
that database on the same server worked. So, the problem seemed to be
local to the database, not the server.

- Jung

--- In firebird-support@yahoogroups.com, "unclejung" <jungv@...> wrote:
>
> Yes, I am aware of the transaction gap problem, having dealt with it
> since Interbase 4 day. I am narrowing down the cause, but it will take
> a little while to find the culprit, as I don't have direct control
> over all the apps that access this database.
>
> Thanks for your suggestions regarding trigger as a cause for infinite
> loop that kill resources. I will look into that.
>
> Our Firebird 1.52 server has been crashing daily and I couldn't
> pinpoint the cause. The problem seemed to have a different outcome
> after upgrading to 2.01. We have other Firebird servers (1.52 and
> 2.01) that have been absolutely reliable, so I am pretty sure we can
> figure this out eventually.
>
> Thanks for your suggestions.
>
> - Jung
>
> --- In firebird-support@yahoogroups.com, Helen Borrie <helebor@> wrote:
> >
> > At 10:54 AM 8/08/2007, you wrote:
> > >I just upgraded on of our server from from Firebird 1.52 to 2.01 on
> > >Windows 2000. I am encountering this error after the server was
up for
> > >a few hours:
> > >
> > >Request depth exceeded. (Recursive definition?)
> > >
> > >When the server was in this state, simple queries will work, but
> > >simple store procedures won't, returning the above error. Restarting
> > >the Firebird server service will get rid of the problem until it
> > >happens again in a few hours.
> > >
> > >The other problem I am dealing with is the large gap between the
> > >oldest active and the next transaction:
> > >
> > >ODS version 11.0
> > >Oldest transaction 4924198
> > >Oldest active 4924199
> > >Oldest snapshot 4924199
> > >Next transaction 5063357
> > >
> > >This was a problem before I upgraded to 2.01, but could it be what's
> > >killing the server in 2.01 and 1.52 some how was able to handle this
> > >better.
> >
> > No, a garbage gap of ~ 140,000 would tend to slow things down in 2.01
> > in the same way as it did in 1.52.
> >
> > You have an operation somewhere that is in an endless loop. If it's
> > just allowed to go on unheeded, eventually it will exhaust resources
> > and the server will die. In the meantime, that operation is highly
> > likely to be contributing generously to the garbage gap and
> > inhibiting garbage collection the way you report.
> >
> > Have a look at triggers that perform operations on other tables, or
> > worse, on their own table or a dependent table. That's a common
> > source of recursion and it's a design problem that won't go away on
> its own.
> >
> > ./heLen
> >
> >
> >
> > >Help appreciated,
> > >
> > >- Jung
> > >
> > >
> > >
> > >++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > >
> > >Visit http://www.firebirdsql.org and click the Resources item
> > >on the main (top) menu. Try Knowledgebase and FAQ links !
> > >
> > >Also search the knowledgebases at http://www.ibphoenix.com
> > >
> > >++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > >
> > >Yahoo! Groups Links
> > >
> > >
> > >
> >
>