11g R2 Grid Infrastructure asmca cannot detect 11g R1 ASM instance

Question

I am playing with 11g ASM with XEN (one node on host (or dom0) and another node on vm (or dom1)) and was trying to do a Clusterware + ASM upgrade from 11.0.6 to 11.2.0. After solving some stupid problem, (see [rootupgrade.sh failed on 11g R1 to R2 Upgrade |http://forums.oracle.com/forums/thread.jspa?threadID=1005424&tstart=0]) so that all services can be started in both nodes, I am trying to upgrade ASM.
I am trying asmca. But it is unable to detect the 11g R1 ASM instance. When started in GUI mode, it just offer to create ASM. When started in silent mode for upgrade, it shows:
oracle@penguin-dom0:~> asmca -silent -upgradeASM
Enter ASMSNMP Password
ORA-01034: ORACLE not availableAnd I can't find anything in ASM alert log
ASM is running from 11g R1 home
oracle@penguin-dom0:~> ps -ef|grep asm
oracle   21811     1  0 Jan06 ?        00:00:00 asm_pmon_+ASM1
oracle   21813     1  0 Jan06 ?        00:00:03 asm_vktm_+ASM1
oracle   21817     1  0 Jan06 ?        00:00:02 asm_diag_+ASM1
oracle   21819     1  0 Jan06 ?        00:00:00 asm_ping_+ASM1
oracle   21821     1  0 Jan06 ?        00:00:00 asm_psp0_+ASM1
oracle   21825     1  0 Jan06 ?        00:00:09 asm_dia0_+ASM1
oracle   21827     1  0 Jan06 ?        00:00:05 asm_lmon_+ASM1
oracle   21829     1  0 Jan06 ?        00:00:03 asm_lmd0_+ASM1
oracle   21831     1  0 Jan06 ?        00:00:03 asm_lms0_+ASM1
oracle   21835     1  0 Jan06 ?        00:00:00 asm_mman_+ASM1
oracle   21837     1  0 Jan06 ?        00:00:00 asm_dbw0_+ASM1
oracle   21839     1  0 Jan06 ?        00:00:00 asm_lgwr_+ASM1
oracle   21841     1  0 Jan06 ?        00:00:00 asm_ckpt_+ASM1
oracle   21843     1  0 Jan06 ?        00:00:00 asm_smon_+ASM1
oracle   21845     1  0 Jan06 ?        00:00:01 asm_rbal_+ASM1
oracle   21847     1  0 Jan06 ?        00:00:00 asm_gmon_+ASM1
oracle   21856     1  0 Jan06 ?        00:00:00 asm_lck0_+ASM1
oracle   22021     1  0 Jan06 ?        00:00:00 /u01/app/oradb/product/11.1.0/db_1/bin/racgimon daemon ora.penguin-dom0.ASM1.asm11g R1 Listener is serving to ASM
oracle@penguin-dom0:~> lsnrctl status
LSNRCTL for Linux: Version 11.1.0.6.0 - Production on 07-JAN-2010 00:33:44
Copyright (c) 1991, 2007, Oracle.  All rights reserved.
Connecting to (ADDRESS=(PROTOCOL=tcp)(HOST=)(PORT=1521))
STATUS of the LISTENER
Alias                     LISTENER_PENGUIN-DOM0
Version                   TNSLSNR for Linux: Version 11.1.0.6.0 - Production
Start Date                06-JAN-2010 22:29:51
Uptime                    0 days 2 hr. 3 min. 53 sec
Trace Level               off
Security                  ON: Local OS Authentication
SNMP                      OFF
Listener Parameter File   /u01/app/oradb/product/11.1.0/db_1/network/admin/listener.ora
Listener Log File         /u01/app/oradb/diag/tnslsnr/penguin-dom0/listener_penguin-dom0/alert/log.xml
Listening Endpoints Summary...
  (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.2.201)(PORT=1521)))
  (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.2.101)(PORT=1521)))
Services Summary...
Service "+ASM" has 1 instance(s).
  Instance "+ASM1", status READY, has 1 handler(s) for this service...
Service "+ASM_XPT" has 1 instance(s).
  Instance "+ASM1", status READY, has 1 handler(s) for this service...
The command completed successfullyNo tnsnames.ora in 11g R1 or Grid Infrastructure home.
Setting ORACLE_HOME to 11g R1 home and ORACLE_SID=+ASM1, I can connect to ASM instance as sysasm using R1 sqlplus
oracle@penguin-dom0:~> export ORACLE_HOME=/u01/app/oradb/product/11.1.0/db_1
oracle@penguin-dom0:~> export ORACLE_SID=+ASM1
oracle@penguin-dom0:~> $ORACLE_HOME/bin/sqlplus / as sysasm
SQL*Plus: Release 11.1.0.6.0 - Production on Thu Jan 7 00:40:15 2010
Copyright (c) 1982, 2007, Oracle.  All rights reserved.
Connected to:
Oracle Database 11g Enterprise Edition Release 11.1.0.6.0 - 64bit Production
With the Partitioning, Real Application Clusters, Oracle Label Security, OLAP,
Data Mining and Real Application Testing options
SQL> Tried different combinations of ORACLE_HOME (11g R1 db home, grid infrastructure home, 11g R1 clusterware home) and ORACLE_SID (+ASM, +ASM1), but asmca still cannot connect to ASM instance.
Some documents mentioned to configure 'Oracle Restart' and the ASM instance name is +ASM. Are these necessary requirements? If so, where can I find information on Oracle Restart and how to rename the ASM instance? Besides, the documents said it is recommanded to upgrade with OUI, but I've already installed the software, can I still use OUI?
OS: OpenSUSE 11.1, XEN DOM0 and DOM1
Oracle: 11.0.6 upgrade to 11.2.0
Edited by: thtsang on Jan 6, 2010 8:47 AM
Edited by: thtsang on Jan 6, 2010 5:33 PM

Answer 1

Hi Thtsang,
sorry for not coming back to you earlier and thanks for "the helpful".
So, are you saying that after you have removed all the environment variables, you still got the error message "ORA-01034: ORACLE not available"? I thought you got an "ASM upgrade failed", because ASM was down at the time of the upgrade, which you then corrected saying: "In fact, ASM instance was started from Grid Infrastructure home.". Overall, I need to say that I am a bit confused to be honest.
The alert log entry sounds promising, although, one would need to see more context. For: "But when I use crsctl / srvctl to start ASM, it is still running in 11g R1 Oracle Home." So, something went wrong with the update. For: "BTW, if asmca does not work for me, can I upgrade manually?" - ASMCA is the tool that will perform an ASM upgrade. You can call it interactively, but you will have to use it.
Support note ID 950200.1 - "ASM 11gR2: How To Upgrade An ASM Instance To Release 11gR2 (11.2.0.1.0)?" describes an ASM upgrade to 11.2. Maybe this note can help you and if it is only to analyze what stage your cluster is in right now. It sounds like, this would be first thing that we need to find out.
Good luck. Thanks,
Markus