Re: Oracle screws up Sun host

Comment inline

Rick Denoire <100.17706_at_germanynet.de> schreef in berichtnieuws 0tt8avsc17fb3dov5qs2756ipllb693rsr_at_4ax.com...
| "Anton Buijs" <remove_aammbuijs_at_xs4all.nl> wrote:
|
|
| >Try to use shutdown abort. I don't see a reason for removing the lk<sid>
| >file.
|
| If you can't connect in the first place, you won't be able to use any
| command.

You can't connect to the Sun box? Or because svrmgr connect internal does not work? That would be very strange
You mention in another reply that you don't want to use shutdown abort. I use it all the time but in a special way. Before db backup we do shutdown abort, startup restrict and shutdown immediate. Shutdown immediate can hang (when a system call of a session does not finish for some reason), causing the backup to fail, and we don't want that to happen. (this will give some reactions).

| >Why do you conclude "the host is screwed up" and needs a reboot?
|
| I have found some hints fixing metaphore problems and the like, but
| they only work save if there is only one instance in the host.

You mean semaphores are not removed? And you can't see which are belonging to the other instance?
Thought that was your problem. Can't test or help here, don't have a Sun box at hand.

| >Further I would certainly look for the trace files produced by the
ORA-600
| >errors. There is a utility on Metalink where you can load the dump stack
and
| >get possible known problems reported.
|
| Is it a utility that one can download or is it an online feature on
| Metalink? But that is a good idea anyway.

It's an online utility, not obvious to find: click "Top tech docs" in the menu bar on the left. Choose "Database". Under "Oracle Internals" you find "Oracle Internal Errors". Then "Getting Started" and finally: "Ora-600 lookup" Copy/paste the stack trace from the trace file in that window and you will find quickly if your problem is known. Received on Tue Apr 22 2003 - 02:45:50 CDT