4108GL problems

[Paul Clayton]

My configuration is as follows. I have a 5308XL in the core where all my major servers connect. Two fibre links connect to a 4108GL using LACP. Another two fibre links connecting to 4 cisco switches, using STP.
My problem. Every day the 4108 appears to die. We cannot ping or telnet or HTTP into the switch. Sniffing the network port, shows traffic still passing, and I can ping the other switches with no problem for a while, then connections drop. What I have noticed, is that the event appears to triggered by a colleague who starts a browser connection to the switch. Left for about 5 minutes the switch will recover.

[Mario Gauthier]

I think I had the same problem before.

How did you wire your two fiber links between the 5308 and 4108. I mean two ports on same card onthe 4108 or one above the other one using two 3 slots cards.

[Mario Gauthier]

There is a issu with trunking on thoses switches. The backplane does not permit two ports on different card. When I set up the trunk on the same card I resolved the problem. I mean A1 and A2 or C1 and C2 for example.We came up with this solution with a expert and talking with HP engineer when we look more closely what was happening with packets who contains the vlan ID.

[Arimo Laine_2]

Hi.

Just FYI: Current 41XX FW rev G.07.26 has a SW fix for connectivity problems in trunks spanning over different modules.

http://www.hp.com/rnd/software/switches.htm

HTH,
~A~

[Mario Gauthier]

You're right Arimo, they problably fix it with this software upgrade.

Thanks