Configure Nagios for Oracle

Hi,

I have recently configured nagios for monitoring some of my production servers and it works fine but now I need to monitor my Oracle RAC servers.

I have downloaded the following plugin (check_oracle-cluster) and have installed it on one of my RAC servers.

Nagios Exchange - Check ORACLE CLUSTER "RAC"

My problem is configuring it to work correctly. I was unable to obtain any information on this?

[root@mvdbs1 libexec]# ./check_oracle_cluster -n ddnode1           
check_oracle_cluster Version 0.81
usage: check_oracle_cluster -n [options]
Global Options:
 -n -- node
Options:
 -a -- ASM
 -l -- LISTENER
 -g -- GSD
 -o -- ONS
 -v -- VIP
Ex.
check_oracle_cluster -n dbnode1 -a
[root@mvdbs1 libexec]# 
[root@mvdbs1 libexec]# ./check_oracle_cluster -n dbnode1 -l
ORACLE_CLUSTER_lsnr WARNING: UNKNOW problem(s)

If someone has experienced this before and can advise?

R,
D.

try

./check_oracle_cluster -n dbnode1 -a

can you provide the output of

cat /etc/ocfs2/cluster.conf

i am trying to learn RAC cluster

Hi,

[root@mvdbs1 libexec]# ./check_oracle_cluster -n mvdbs1 -a
ORACLE_CLUSTER_asm WARNING: UNKNOW problem(s)
[root@mvdbs1 libexec]# cat /etc/ocfs2/cluster.conf 
node:
        ip_port = 7777
        ip_address = <IP_ADDR of host>
        number = 0
        name = MVDBS1
        cluster = ocfs2
node:
        ip_port = 7777
        ip_address = <IP_ADDR of host>
        number = 1
        name = MVDBS2
        cluster = ocfs2
cluster:
        node_count = 2
        name = ocfs2

check the access of asm disks from two nodes

try

./check_oracle_cluster -n mvdbs2 -a

try different options passing node 1 & 2 alternatively

cluster commands are not same for all clusters

Just a stab in the dark here, but have you tried ./check_oracle_cluster -n MVDBS1 -a instead? From what I read in that plugin it does a case-sensitive check of the output of some Oracle command, and I know Oracle can be pretty anal about case.

Hi,

Robo - tried all possible options with same result.

Pludi - tried both lower/up case options with same result.

R,
D.

Alright, lets see:

• Did you put utils.sh in the same path as the check_oracle_cluster script?
• Have you checked if the CRSSTAT_PATH variable in the script, line 12, points to the correct executable, both for sudo and for crsstat?
• Have you tried removing the sudo call when running as root?
• Have you checked that crsstat is running correctly?

Hi,

• Both uyils.sh & check_oracle_cluster are in the same location (....libexec/)
• the crsstat path variable is pointing to the correct path location (same as above)
• I have removed the sudo call as I am running this from root
• The crsstat is NOT running locally. I am running the check_oracle_cluster scripts from my nagios server (not the db server). I have however got my oracle client installed.

R,
D.