CSS Daemon error !!

Question

Hello ,
I am having problems connecting from RMAN to the database (Oracle 10.2.0.3 running on CRS , AIX 5.3).
When i am trying to connect it shows following errors.
pr:/u00/oracle/product/10.2.0/crs/bin%*> rman target /
Recovery Manager: Release 10.2.0.3.0 - Production on Mon Oct 13 17:09:06 2008
Copyright (c) 1982, 2005, Oracle. All rights reserved.
RMAN-06900: WARNING: unable to generate V$RMAN_STATUS or V$RMAN_OUTPUT row
RMAN-06901: WARNING: disabling update of the V$RMAN_STATUS and V$RMAN_OUTPUT rows
ORACLE error from target database:
ORA-29701: unable to connect to Cluster Manager
connected to target database: PSP (DBID=695888555)
== Then i checked CSSD which shows like below
pr://u00/oracle/product/10.2.0/crs/bin:/crsctl check crs
Failure 1 contacting CSS daemon
ON 08th Oct , system admin added one more port and i think they cloned from existing (PR) to (PR-dr) from Smitty.. I think that time CRS switched from PR to PR-dr while system admins changed back the configuration.
pr:/u00/oracle/product/10.2.0/crs/log%*> ls -lrt
total 0
drwxrwx--- 2 oracle dba 256 Mar 06 2008 crs
drwxr-xr-t 8 root dba 256 Mar 06 2008 pr
drwxr-xr-x 3 root system 256 Oct 08 14:00 pr-dr
pr:/u00/oracle/product/10.2.0/crs/log/PR-dr/racg ls -lrt
drwxr-xr-x 2 root system 256 Oct 08 14:00 racgmain
drwxr-xr-x 2 root system 256 Oct 08 14:00 racgeut
-rw-r--r-- 1 root system 1737625 Oct 13 17:15 ora.pr.vip.log
--- tail ora.pra.vip.log
2008-10-13 17:13:53.231: [    RACG][1] [802942][1][ora.pr.vip]: clsrccssgetctx: clsssinit() failed. rc=21
2008-10-13 17:13:53.264: [    RACG][1] [802942][1][ora.pr.vip]: clsrcgetprsrctx: prsr_init_ext returned rc = 5
I think if i will stop the crs on this node while crs will be active on other node .. Will this work ?
Or can i use crsstart to restart cssd
Or i can use localconfig
Please help.
Thanks
Edited by: P explorer on Oct 13, 2008 5:22 PM

Answer 1

Hi P explorer
Sorry for the late answer, may this work-around should help you out of that issue. it applies to Version: 10.1.0.2 to 10.2.0.4
Cause
The hidden directory '/var/tmp/.oracle' was removed while instances & the CRS stack were up and running. Typically this directory contains a number of "special" socket files that are used by local clients to connect via the IPC protocol (sqlnet) to various Oracle processes including the TNS listener, the CSS, CRS & EVM daemons or even the database instance. These files are created when the "listening" process starts.
Solution
The only way to re-create these special files is to restart (instance, listener, CRS). In a RAC environment this requires the shutdown & restart of the entire CRS stack.
As these special files are required to communicate with the various CRS daemons, it most likely will not be possible to stop (and restart) the CRS stack using the following commands as user root - but it won't hurt to try it anyway:
10gR1:
# /etc/init.d./init.crs stop
# /etc/init.d./init.crs start
10gR2:
# $ORA_CRS_HOME/bin/crsctl stop crs
# $ORA_CRS_HOME/bin/crsctl start crs
If the above fails to successfully stop the CRS stack, a system reboot will be inevitable.
As for deleting files from temporary directory via a cronjob:
the directory '/var/tmp/.oracle' (on some platform /tmp/.oracle) should be excluded from such jobs. The files in this directory on occupy a few bytes and generally do not need to be cleaned up.
A good practice for deleting files is to check if the file is still being opened by a process, most Unix platforms have a 'fuser' command which returns the process IDs of such processes.
Good luck.