Cluster config value changes

[paabhi]

Hi,

We have a cluster running with SG version 11.16.The parameters "NETWORK_POLLING_INTERVAL","NODE_TIMEOUT " ," QS_POLLING_INTERVAL " etc.We have applied the correct values and started the cluster.It came up without any error.I have started all the packages and DB also started without any error.Upon verification from oracle team , the listener was not starting and the status was "oracle 6144 6127 0 09:03:31 ? 0:00 sh -c /ofa/u001/IEWBDAV/app/oracle/product/10.2.0/bin/lsnrctl start IEWBDAV
oracle 5345 5341 0 09:01:35 ? 0:00 /ofa/u001/IEINTE02/app/oracle/product/8.1.7/bin/lsnrctl start IEINTE02".

Could anyone have any clue.I am confused as 2 dbs running fine on that server and some 10 dbs running fine on the second node.

Abhilash

[melvyn burnard]

and what makes you think those cluster values have any bearing whatsoever on your problem?
Get the DBA to have a look through the Oracle logs etc.

[paabhi]

The DBA is saying all fine.My knowledge says all fine from unix.I have rebooted the server also.Just want to ensure.

[Subhajit Khanbarman]

Could you check the Listener log and see if you have an error in there.

Send me your Listener Entry.

It might be environment related.

Please try and set the follwoing:

export ORACLE_HOME=/user/../app/oracle/product/...

export PATH=$ORACLE_HOME/bin:$PATH

export LD_LIBRARY_PATH=$ORACLE_HOME/lib

ANd then try again...

[paabhi]

Hello,

There is around 14 dbs running and each db is running throu cluster package.Each db is having separate Oracle binaries and oracle home.After all the DB is coming up and i can see the pmon entries in ps -aef

[Basheer_2]

Hey Paabhi.

First the cluster params have nothing to do with starting the Oracle. You are getting " The listener was not starting". So Please chk the status of the listerner using lsnrctl.

This pkg must have 2 files
1) pkg.conf file
2) pkg.cntl

can you start the pkg manually. pkgname start
(/etc/cmcluster/cluster-name/pkgname/pkgname start or pkgname stop)

Logs: (path may be difft for your case)
1) /etc/cmcluster/cluster-name/pkgname/pkg.cntl.log

this will give you all the detials and where it stopped.

[Subhajit Khanbarman]

I think 'pmon' entry will tell you about the database but not about the listener.

If you have multiple listeners for each and every database, you have to run those listeners for each and every DB.

Could you please send the files named tnsnames.ora, listener.ora and the logfile of the listener which is failed to run.

Please send the output for each listener for the following,

ps -ef|grep tns or,
lsnrctl status
lsnrctl services
tnsping

# lsof -i:1521
or
# netstat -an | awk '{if(z[split($4,z,".")]==1521) print $0}'

Also check in listener log file whether you are getting any of the following errors,

TNS-12545: Connect failed because target host or object does not exist
TNS-12560: TNS:protocol adapter error
TNS-00515: Connect failed because target host or object does not exist
HPUX Error: 13: Permission denied

or more information see also:
http://download-west.oracle.com/docs/cd/B10501_01/network.920/a96580/listener.htm#454524