Subject Re: [firebird-support] Re: event log error 251
Author Robert martin
> Rik,
> I suspected that it was just an information message too, but had not
> any references to back it up. IMO, the real issue the original poster
> must concern themselves with is why guardian needed to restart the
> server in the first place.
> All guardian does is restart the Firebird service if it crashes for
> whatever reason. On any modern Windows version, you don't actually
> need it because all of the recovery options are available in the
> services mmc pages.
> I have seen 3 causes for database server processes crashing.
> * Faulty UDFs
> With Superserver, UDFs must be threadsafe. Silly bugs like not using
> the right calling convention can also cause troubles.
> * Corrupt database
> A corrupt database can cause the engine to panic and shutdown when the
> record version with the corruption is read.
> * Faulty Hardware
> If the machine itself is suspect, this can cause the engine to panic
> and close.
> The first place to check is the Firebird log. Forget the guardian
> restarting, we need to get to the bottom of why it needed to be restarted.
> Adam

Thanks Rock and Adam. I didn't really know very much about the services

Adam really appreciate your 3 possible causes :) I think you might have
found the issue for me.

1. We don't use any extra UDFs
2. Database is not corrupt (I think).
3. Machine has been running for months and seems reasonably stable.

However looking in the firebird log (now why didn't I think of that !)
shows the following message (a number of times)...

SHSERVER (Client) Fri May 19 15:24:48 2006
Control services error 1061

SHSERVER (Client) Fri May 19 15:24:54 2006
Guardian starting: C:\Program

So now I need to know, what causes a 'Control services error 1061'?

Rob Martin
Software Engineer

phone +64 03 377 0495
fax +64 03 377 0496

Wild Software Ltd