a fresh SI 184.108.40.206 database on ASM installation in Oracle restart shows a strange behaviour over shutdown via crsctl stop has.
The database is configured for a shutdown immediate, of course:
+$srvctl config database -d ORCL | grep Stop+ Stop options: immediate
but stopping by
+$crsctl stop has+
the Oracle Restart stack produces a shutdown abort..
I was unable to find official documentation about that (expect for crsctl stop crs which causes a ASM shutdown abort , that is in RAC case. )
Is it an expected behaviour or is something like a bug?
yes this is expecting behaviour. So you can ignore it. For same issue I prefer to close db first than give crsctl stop crs command on both nodes.
By the way I suggest for your future CRS issue use Forum Home » High Availability » RAC, ASM & Clusterware Installation which is RAC dedicated forum site.
Thank for you reply, but this is an Oracle Restart Issue,
that is not RAC but single instance.
Besides, as when you type crsctl stop has ASM instance is closed cleanly (by a shutdown immediate),
it seems strange to me that database is not.
Yes , thanks it explains crsctl stop commands should be
used as emergency procedure to stop ASM or databse and that a shutdown immediate on a database could provide an unwanted delay.
Anyway, as both database and ASM are Oracle Restart managed resources it's pretty strange that
while asm shutdown is fine , database is not.