Cluster Issue

[Ezzat_1]

we have a cluster on rp7420 , its 2 node cluster. The problem is whenever the network fals for 1 min, the cluster goes down and yesterday it tooks about 8 hours for the cluster to startup.

[melvyn burnard]

8 hours?
I would suggest you check your configuration settings for teh cluster and also explain a bit more about "the network went down"
Did the servers TOC?
Some more detail would help

[Ralph Grothe]

Maybe they have set RUN_SCRIPT_TIMEOUT to 28800
;-)

[Ezzat_1]

No the time-out=300s , the network is having a problem, from time to time. Here is some of the logfile:



Error: Non-uniform connections detected,

Mar 1 18:28:26 mdo1 CM-CMD[2154]: kbs1 lan0 successfully received from mdo1 lan3

Mar 1 18:28:26 mdo1 CM-CMD[2154]: but yakmd01 lan3 did not receive from kbs1 lan0.

Mar 1 18:28:24 ndo1 su: + tty?? root-oracle

Mar 1 18:28:26 ndo1 CM-CMD[2154]: This could be due to heavy network traffic, or heavy load on kbs1.

Mar 1 18:28:26 mdo1 CM-CMD[2154]: kbs1 lan1 successfully received from mdo1 lan3

Mar 1 18:28:26 mdo1 CM-CMD[2154]: but yakmd01 lan3 did not receive from kbs1

[melvyn burnard]

attach your cluster configuration file, also logs from when cluster went down.