Switches, Hubs, and Modems
cancel
Showing results for 
Search instead for 
Did you mean: 

HP4108GL and IGMP problem

Darren Coleman
Occasional Visitor

HP4108GL and IGMP problem

I have a HP4108GL switch which is the root switch in a dual star topology network connected back to a CISCO Catalyst 6513 switch/router. Every so often the switch goes through an IGMP querier election process causing a spike in the processor CPU.
This causes loss of connectivity to the switch for approx 5 minutes but the switch also stops forwarding and processing RSTP BPDU's. This then creates a loop as the other switches seemingly detect a indirect failure a bring the redundant backup links from Blocking to a Forwarding state. Now the backup links have been disconnected to prevent more loops happening but the problem still persists. Even though connectivity is lost to the switch it still forwards packets and all 11 4108's switch's behind this are visible and functioning correctly.
The switch is running version 7.22 and the chassis has been replaced to no avail. Multicast routing is enabled on the CISCO so this querier election should not be happening anyway.
Any suggestions would be welcome.

Cheers

Darren

1 REPLY

Re: HP4108GL and IGMP problem

Darren,

i've seen the same kind of problem with HP2524 causing a crash data message: Software exception at infMcasts.c:894 -- in 'eDrvPoll', task ID = 0xdfa864
-> ASIC IGMP address not found

If this is due to a spike i dont know but it might be. If you set a lower IP address on the Catalyst it will be the only querier, if its still problem you can always unconfig IP address on the 4108 so it CANT be querier.

/Johan
Johan