From Newsgroup: comp.databases.ingres

On Tuesday, March 3, 2020 at 8:42:08 PM UTC+11, jpz...@gmail.com wrote:

After killing processes don't forget to clean memory segments.

And if your Ingres installation fail to stop gracefully and you need to kill processes and clean memory, once it start again and recovery finish his job without pain, don't forget to stop/start Ingres again.

Thanks Jean-Pierre, your advice came in handy last night.

Thanks
Steve
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.databases.ingres

Steve wrote:

On Tuesday, March 3, 2020 at 8:42:08 PM UTC+11, jpz...@gmail.com wrote:

And if your Ingres installation fail to stop gracefully and you need to
kill processes and clean memory, once it start again and recovery finish
his job without pain, don't forget to stop/start Ingres again.

Thanks Jean-Pierre, your advice came in handy last night.

That suggestion is harmless but I refuse to believe it is useful.

If you don't trust the undo/redo recovery to work properly you are
either being superstitious or you have some incidents you should have
reported to tech support to be fixed properly.

I'm going with superstitious. Ingres is not fragile.

Roy
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.databases.ingres

I've had to kill ingres processes a few times in my career.
Just check the last few lines of iircp.log to make sure everything cleaned up ok.
I remember there was a time when the recovery server lost access to the transaction log a few years ago due to disk failure. Good times :)
Andro Laframboise
Conseiller Principal Base de Donnoes, Direction gonorale de la gestion de l'information et Technologie de l'information
Biblioth*que et Archives Canada / Gouvernement du Canada andre.laframboise@canada.ca / Tol. : 613-298-1346
Senior Database Advisor, Information Management and Information Technology Branch
Library and Archives Canada / Government of Canada
andre.laframboise@canada.ca / Tel: 613-298-1346
-----Original Message-----
From: info-ingres-bounces@lists.planetingres.org <info-ingres-bounces@lists.planetingres.org> On Behalf Of Roy Hann
Sent: Tuesday, November 17, 2020 9:38 AM
To: info-ingres@lists.planetingres.org
Subject: Re: [Info-ingres] Waiting indefinitely for the logging system to shut down
Steve wrote:

On Tuesday, March 3, 2020 at 8:42:08 PM UTC+11, jpz...@gmail.com wrote:

And if your Ingres installation fail to stop gracefully and you need
to kill processes and clean memory, once it start again and recovery
finish his job without pain, don't forget to stop/start Ingres again.

Thanks Jean-Pierre, your advice came in handy last night.

That suggestion is harmless but I refuse to believe it is useful.
If you don't trust the undo/redo recovery to work properly you are either being superstitious or you have some incidents you should have reported to tech support to be fixed properly.
I'm going with superstitious. Ingres is not fragile.
Roy
_______________________________________________
Info-ingres mailing list
Info-ingres@lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.databases.ingres

On Wednesday, November 18, 2020 at 1:38:25 AM UTC+11, Roy Hann wrote:

Steve wrote:

On Tuesday, March 3, 2020 at 8:42:08 PM UTC+11, jpz...@gmail.com wrote:

And if your Ingres installation fail to stop gracefully and you need to >> kill processes and clean memory, once it start again and recovery finish >> his job without pain, don't forget to stop/start Ingres again.

Thanks Jean-Pierre, your advice came in handy last night.

That suggestion is harmless but I refuse to believe it is useful.

If you don't trust the undo/redo recovery to work properly you are
either being superstitious or you have some incidents you should have reported to tech support to be fixed properly.

I'm going with superstitious. Ingres is not fragile.

Roy

For my situation, Ingres continued to fail to shutdown and only did so gracefully after I ran ipcclean, after killing it.
That said, you're right Roy, there was something up with the database whose transactions the recovery process was supposedly working through. Even when Ingres was up and running, I could not open a session to that database, so something went awry at some point. This was our DR environment and that particular database is static, so I could simply recovery it from its latest production checkpoint.
Steve
--- Synchronet 3.21b-Linux NewsLink 1.2