RE: Now I've seen it all
Posted in 2004
Malcolm,
If Informix is unable to start and crashes after
an attempt to make the first 'fork' (start ADM VP) on Linux,
it is most probably caused by incorrect LIBPTHREAD library.
Libpthread.so is a component of the GLIBC package.
The workaround is to install another i686 GLIBC package
(or at least another libpthread library)
------------------------------------------
Alexey Sonkin
> -----Original Message-----
> From: malcolm weallans [mailto:malcolm.iiug@btopenworld.com]
> Sent: Tuesday, July 13, 2004 12:27 PM
> To: informix-list@iiug.org
> Subject: Now I've seen it all
>
> Today we were investigating why our database engine
> would not restart following a relocation. The system
> is IDS 9.30.UC5 running on SUSE Linux SLES 8.
>
> Here's the good bit.
>
> We couldn't start the engine using oninit so we used
> oninit -v. The engine did not succeed in starting the> ADM VP. There were no messages in the message log,
> the shared memory had been created, but there were no
> oninit processes running. We tried various things
> such as starting oninit as root, rebooting the wholde
> machine, and even resorted to praying to the Informix
> Handbook, but nothing worked. Then we tried starting
> up using onmonitor and ....
>
> it worked?
>
> Now I'm really baffled. What can I possibly have
> missed? We've checked raw-device permissions, we've
> checked /etc/hosts and everything looks OK.
>
> Does anybody have any ideas of what we might have
> missed?
>
> regards
>
> Malcolm
> sending to informix-list
sending to informix-list