Broadcasts caused by Baystack 420
Bookmark this page:
 Yahoo!
 Google
 Windows Live
 del.icio.us
 digg
 Netscape
|
|
|||||||||||||
|
Posted by SF on August 11, 2006, 2:43 am
Please log in for more thread options i have two Baystack 420 in a stack running. There are 15 VLANs configured and in the 15th are all ports members. But only the ports my servers are connected to have the PVID 15. With regard to the manual this is the correct way of sharing central resources with all VLANs. When a frame enters the switch trough a server port with a unique computer as target, the frame is being broadcastet throughout all switchports even the mac-address table of the switches and arp-caches of the servers are correct. A network sniffer confirms that the mac-address of source and target is written in the frame header. The switches are running with FW-Version 3.1.0.5 and SW-Version 3.1.0.04 - I cant get Nortel Support because we don't have a valid service contract. I would be grateful if anybody knows how to solve this problem. Thanks in advance Stefan | |||||||||||||
|
Posted by lou on August 18, 2006, 6:01 pm
Please log in for more thread options if the mac destined to an unknown station to 420, of course it'll treat it as an unknown unicast, and thus will send the pkt to all ports belong to vid 15 (which happens to have all ports in it). perhaps you do have a router somewhere attached to this 420? and hopefully you do turn on 802.1q/tagging on 420 uplink? good luck. SF wrote: | |||||||||||||
|
Posted by SF on August 25, 2006, 1:37 am
Please log in for more thread options Thanks for your answer,
the thing is that the target station is not unknown to the switch. It is correctly shown in the mac address table of the switch and frames from the station to the server ports are send as unicasts, as they are intended to. Only the answer frame from the server with a station connected to the second stack unit (servers are connected to the first unit) is being broadcasted inspite of correctly addressed frame headers. | |||||||||||||
| Similar Threads | Posted |
| Broadcasts caused by Baystack 420 | August 11, 2006, 2:43 am |
| Exchange Lines locking up on Norstar Modular caused by Startalk?? | July 1, 2005, 2:39 am |
| Baystack 450 & 470 | July 18, 2005, 2:44 am |
| Baystack 460 24T Pwr | January 8, 2006, 10:04 pm |
| Pb with baystack 450 | September 12, 2007, 4:57 am |
| baystack 450-24t 470-48t | March 2, 2008, 8:32 am |
| Baystack 470 not booting | March 4, 2005, 5:11 pm |
| Radius & Baystack 450 | May 17, 2005, 4:16 pm |
| baystack 425 switches | July 15, 2005, 7:30 pm |
| Switch BayStack 310-24T | September 28, 2005, 9:00 pm |
| Baystack 470 Stack | July 8, 2005, 4:44 pm |
| Baystack 450 old firmware | February 17, 2006, 7:34 am |
| Baystack 350 SNMP | February 18, 2006, 12:00 pm |
| BayStack 350-24T Reset | December 21, 2005, 3:33 pm |
| Baystack 470 QoS Configuration | January 23, 2007, 10:16 am |
>
> i have two Baystack 420 in a stack running. There are 15 VLANs
> configured and in the 15th are all ports members. But only the ports my
> servers are connected to have the PVID 15. With regard to the manual
> this is the correct way of sharing central resources with all VLANs.
>
> When a frame enters the switch trough a server port with a unique
> computer as target, the frame is being broadcastet throughout all
> switchports even the mac-address table of the switches and arp-caches
> of the servers are correct. A network sniffer confirms that the
> mac-address of source and target is written in the frame header.
>
> The switches are running with FW-Version 3.1.0.5 and SW-Version
> 3.1.0.04 - I cant get Nortel Support because we don't have a valid
> service contract. I would be grateful if anybody knows how to solve
> this problem.
>
> Thanks in advance
>
> Stefan