freeipmi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Freeipmi-devel] Re: BMC/ARP investigation update


From: Anand Babu
Subject: Re: [Freeipmi-devel] Re: BMC/ARP investigation update
Date: Thu, 05 Feb 2004 18:02:44 -0800
User-agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.3 (gnu/linux)

Hi Al,
I noticed PortFast configuration between switches. This can also cause
trouble.

SOME THEORY
-----------
PortFast
Spanning Tree PortFast causes an interface configured as a Layer 2
access port to enter the forwarding state immediately, bypassing the
listening and learning states. You can use PortFast on Layer 2 access
ports connected to a single workstation or server to allow those
devices to connect to the network immediately, rather than waiting for
spanning tree to converge. If the interface receives a bridge protocol
data unit (BPDU), which should not happen if the interface is
connected to a single workstation or server, spanning tree puts the
port into the blocking state.


     Note   Because the purpose of PortFast is to minimize the time
     access ports must wait for spanning tree to converge, it is most
     effective when used on access ports. If you enable PortFast on a
     port connecting to another switch, you risk creating a spanning
     tree loop.


,----[ Anand Babu <address@hidden> ]
| Cause of this bug is - Intel GigE goes into some confused state upon
| "suspend" which affects Ethernet switching.
| 
| Effects of a cross-over cable or a Ethernet hub are the same. Packets
| reach every port blindly (no switching).
| 
| There is a configuration option in Cisco IOS to make a switch behave
| like hub when there is no matching ARP entry or during the
| spanning-tree table-rebuild phase.
| 
| 
| I think Ben should take a look at rmmod issue. I have noticed this
| problem with Redhat's default kernel (without elan) too.
| 
| -ab
`----

,----[ Albert Chu <address@hidden> ]
| 1) Why would we be able to power control when I hook up my laptop back
| to back against a halted node??
| 
| 2) Why would packet drops to a halted node be 90% and not 100%?
| 
| 3) Why did the power control/hatl problem not occur on our test
| cluster with a cisco 3550 switch?
| 
| 4) Why did "rmmod e1000" not occur on our test cluster??
`----

-- 
Anand Babu
Free as in Freedom <www.gnu.org>


_______________________________________________
Freeipmi-devel mailing list
address@hidden
http://mail.nongnu.org/mailman/listinfo/freeipmi-devel


-- 
 _.|_ 
(_||_)
Free as in Freedom <www.gnu.org>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]