lsnrctl can start listener, crs_start cannot

I'm in the process of setting up a two-node Oracle 10.2.0.3 RAC cluster on SLES 10. I ran a clusterware installation and two database software installations (db_1 to be used for the databases, and db_2 to be used for asm). The database software installations were run with the "install database software only" option.

Prior to setting up asm, I ran netca to define the listeners, but netca gives me pop-up warnings:

Problem in configuration: ol01: LISTENER_OL01 Listener start failed. Listener may already be running. CRS-0215: Could not start resource 'ora.ol01.LISTENER_OL01.lsnr'.

Problem in configuration: ol02: LISTENER_OL02 Listener start failed. Listener may already be running. CRS-0215: Could not start resource 'ora.ol02.LISTENER_OL02.lsnr'.

I verified that the listeners are not running on either node.

'oerr CRS 215' returns
215, 0, "Could not start resource '%s'." // *Cause: There was an internal error while starting the resource. // *Action: Check the CRS daemon log file.

but the daemon log is not exactly helpful: 2007-06-07 12:01:24.134: [ CRSRES][1411561792]0Attempting to start
`ora.ol01.LI

STENER_OL01.lsnr` on member `ol01`
2007-06-07 12:01:24.623: [ CRSAPP][1411561792]0StartResource error for ora.ol01
.LISTENER_OL01.lsnr error code = 1
2007-06-07 12:01:24.928: [ CRSRES][1411561792]0Start of
`ora.ol01.LISTENER_OL01

.lsnr` on member `ol01` failed.

I tried to start the listeners using 'crs_start ora.ol01.LISTENER_OL01.lsnr -c ol01'

but that again fails with
Attempting to start `ora.ol01.LISTENER_OL01.lsnr` on member `ol01` Start of `ora.ol01.LISTENER_OL01.lsnr` on member `ol01` failed. CRS-0215: Could not start resource 'ora.ol01.LISTENER_OL01.lsnr'.

Starting them with lsnrctl works fine, but crs_stat still sees them as offline:

NAME=ora.ol01.LISTENER_OL01.lsnr
TYPE=application
TARGET=ONLINE
STATE=OFFLINE I removed the listeners (crs_unregister), and tried defining them during the asm setup using dbca. Same result.

I found thread
http://groups.google.com/group/comp.databases.oracle.server/browse_thread/thread/55480af8f610bcc4/1eea6d12bab2464a?lnk=gst&q=CRS-0215&rnum=3#1eea6d12bab2464a which discusses a similar error (similar in that the STATE there is UNKNOWN), and but none of the suggestions seem to work.

I ran cluvfy, which only fails on 'Checking system requirements', with a ' No checks registered for this product', but that is 'normal' for SLES10. crs_stat only shows the listeners to be OFFLINE. Everything else is ONLINE. I re-installed the cluster software dropping the domain names from the hostnames, but thta didn't help either.

Initially, the vip-addresses were only defined in /etc/hosts, so I had them added to dns (and verified both the lookup and reverse lookup). Still no luck.

Does anyone have any idea what else I could check?

I'll probably open an SR with Oracle later today, but given that they only just started working on a severity 2 call I opened on May 22nd, I was hoping to get this resolved a bit quicker with your help.

Kind regards,

Herta Received on Thu Jun 07 2007 - 05:59:57 CDT