Unable to run cmquerycl successfully on previously built 4-node cluster

[SuperDome_1]

On running the command
cmquerycl -n atlhp1 -n atlhp2 -n atlhp3 -n atlhp4 -v -C /etc/cmcluster/cluster.asc

I get the error :

Begin checking the nodes..
Looking for other cluster.. Done
Error : Unable to establish communication to node atlhp3
Failed to gather configuration information

There are link lights on the Hyperfabric interconnect cable from that node to the Hyperfabric switch, and the lanscan command shows the interface is up.
But I can't ping the clic0, clic1 or the heartbeat on that node..
I get the following errors in clic_log in /var/adm

Message receive failed on endpoint 10, errno = 238, peek 2

Any suggestions would be welcome..

[Helen French]

Error number 238 is a "connection time out". You need to check if the card is configure properly with proper drivers. Check if you can ping other networks on the same subnet. Mak sure you have these patches installed too:
PHNE_26551 (for 11.00)
PHNE_27745 (for 11.11)

[SuperDome_1]

Thx for your reply. I don't have the patch mentioned installed.. I also can't ping other nodes on the same network (there are the IPs assigned to the clic0 and clic1 interfaces, which are Hyperfabric interfaces) I can ping them intermittently, but not continously.. as I would like.. I do see link lights on the Hyperfabric switch for all 4 nodes.. I'll apply the patch and go from there.. Are there any network timeout parameters I can set at the OS / Kernel level..

Thx

[melvyn burnard]

Well unless you fix your basc networking issues, you will not get Sg to work correctly.
Start with basics, checking speeds, cables, switch etc.

[SuperDome_1]

Thx again.. A new switch had been placed in the cluster.. which was bad.. Replaced the switch with a Cisco one and I was able to create the cluster successfully..

Thx