Hi all,
I have an SR open on this, but progress seems slow, so I thought I'd put
something here....
Environment:
AWS EC2 instance, m5d.2xlarge (8 VCPUs/32GB RAM)
Oracle 18c (18.3.0) ASM/Grid w/ ASMFD configured, single instance (SIHA),
not RAC
Three Oracle Databases: 2x 12.1.0.2 and 1x 11.2.0.4
Everything works fine, except I can't start/stop/config any DB (12c or 11g)
via srvcrl.
If I start/stop w/ SQL*Plus, everything works great.
If I try start with srvctl, I get errors:
[oracle@miscr2 ~]$ . oraenv
ORACLE_SID = [oracle] ? +ASM
The Oracle base has been set to /u01/app
[oracle@miscr2 ~]$ srvctl start database -d BMPR2PRD
PRCR-1079 : Failed to start resource ora.bmpr2prd.db
CRS-5017: The resource action "ora.bmpr2prd.db start" encountered the
following error:
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3640
Additional information: 1804474595
Process ID: 0
Session ID: 0 Serial number: 0
. For details refer to "(:CLSN00107:)" in
"/u01/app/diag/crs/miscr2/crs/trace/ohasd_oraagent_oracle.trc".
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3640
Additional information: 1804474595
Process ID: 0
Session ID: 0 Serial number: 0
CRS-2674: Start of 'ora.bmpr2prd.db' on 'miscr2' failed
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3640
Additional information: 1804474595
Process ID: 0
Session ID: 0 Serial number: 0
If I look into /u01/app/diag/crs/miscr2/crs/trace/ohasd_oraagent_oracle.trc
as instructed, among many other things, I see:
CRS-5017: The resource action "ora.bmpr2prd.db start" encountered the
following error:
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3640
Additional information: 1804474595
Process ID: 0
Session ID: 0 Serial number: 0
. For details refer to "(:CLSN00107:)" in
"/u01/app/diag/crs/miscr2/crs/trace/ohasd_oraagent_oracle.trc".
2018-09-21 17:41:51.511 : AGFW:34227968: {0:0:21154} sending status msg
[CRS-5017: The resource action "ora.bmpr2prd.db start" encountered the
following error:
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
Linux-x86_64 Error: 2: No such file or directory
Additional information: 3640
Additional information: 1804474595
Process ID: 0
Session ID: 0 Serial number: 0
. For details refer to "(:CLSN00107:)" in
"/u01/app/diag/crs/miscr2/crs/trace/ohasd_oraagent_oracle.trc".
] for start for resource: ora.bmpr2prd.db 1 1
Which doesn't really tell me much. Oracle Support had me turn on srvm
tracing and capture more data. When I did that, I discovered this:
[main] [ 2018-09-21 10:25:56.395 EDT ]
[UnixSystem.dorunRemoteExecCmd:4172] Final unix SSH command:
/u01/app/oracle/product/12.1.0/dbhome_1/bin/osdbagrp -r^M
[Thread-14] [ 2018-09-21 10:25:56.395 EDT ] [StreamReader.run:62] In
StreamReader.run ^M
[main] [ 2018-09-21 10:25:56.395 EDT ] [RuntimeExec.runCommand:294]
runCommand: Waiting for the process^M
[Thread-13] [ 2018-09-21 10:25:56.395 EDT ] [StreamReader.run:62] In
StreamReader.run ^M
[Thread-14] [ 2018-09-21 10:25:56.396 EDT ] [StreamReader.run:66]
ERROR>/u01/app/oracle/product/12.1.0/dbhome_1/bin/osdbagrp -r ^M
[Thread-14] [ 2018-09-21 10:25:56.396 EDT ] [StreamReader.run:66]
ERROR>Usage: /u01/app/oracle/product/12.1.0/dbhome_1/bin/osdbagrp -a | -d |
-o | -b | -g | -k^M
So, according to the trace, it runs the 'osdbagrp' command several times
with different command line arguments. -a, -d, -o, -b, -g, and -k all work
and (appear to) produce valid results.
When it runs 'osdbagrp -r', I get a usage error, because, in 12.1.0.2,
osdbagrp does not support the '-r' argument. The 18c version of the
osdbagrp binary works fine, but that doesn't help me, since this is a
12.1.0.2 database.
Anyone have any insight into what's going on here? If I could just get
this automated startup and shutdown working, I can proceed.
Sadly, there is very little documentation on the 'osdbagrp' command, so,
I'm not sure what all the options are. It seems to be related to returning
the various dba groups: dba,asmadmin,asmoper,oper, etc,etc....
Finally, I checked MOS Doc. ID 337737.1, and both 12.1.0.2 and 11.2.0.4
*are* supported with 18c ASM/GRID.
Help?
-Mark
