Run Cluvfy Debug script : cluv.sh
# cluvfy commands to debug partial up and running CRS stack
# Node1 hract21: CRS stack is not starting:w
# Node2 hract22: CRS stakc is fully up and running
# Note : This scripts assumes we are logged in on the failing Node : hract21
~/CLUVFY/bin/cluvfy comp nodecon -n $Node1,$Node2 -i eth1
~/CLUVFY/bin/cluvfy comp nodecon -n $Node1,$Node2 -i eth2
~/CLUVFY/bin/cluvfy comp nodereach -n $Node1,$Node2
~/CLUVFY/bin/cluvfy comp sys -p crs
# Some of the scripts need a working CRS stack :
# When clusterware is not up and running cluvfy comp healthcheck may report the eror below : these errors are expected
# Cluster Manager Integrity FAILED This test checks the integrity of cluster manager across the cluster nodes.
# Cluster Integrity FAILED This test checks the integrity of the cluster.
# OCR Integrity FAILED This test checks the integrity of OCR across the cluster nodes.
# CRS Integrity FAILED This test checks the integrity of Oracle Clusterware stack across the cluster nodes.
ssh $Node2 ~/CLUVFY/bin/cluvfy comp healthcheck -collect cluster -html
ssh $Node2 ~/CLUVFY/bin/cluvfy comp software -n $Node1,$Node2
# Testing multicast
ssh $Node2 ~/CLUVFY/bin/cluvfy stage -post hwos -n $Node1
ssh $Node2 ~/CLUVFY/bin/cluvfy stage -pre crsinst -n $Node1 -networks eth1:192.168.5.0:PUBLIC/eth2:192.168.2.0:cluster_interconnect -verbose
Pages: Page 1, Page 2, Page 3, Page 4, Page 5, Page 6, Page 7, Page 8, Page 9, Page 10, Page 11, Page 12, Page 13, Page 14, Page 15, Page 16
2 thoughts on “Troubleshooting Clusterware startup problems with DTRACE”
This is very helpful
I really like looking through an article that can make people think.
Also, thank you for permitting me to comment!