[c-nsp] switch interface down but passing traffic

[Gert Doering]

Hi,

got hit by a quite interesting bug in the CBS3020 switches yesterday:

sw2-int.blau#sh int g0/24
GigabitEthernet0/24 is up, line protocol is down (disabled) 
...
  5 minute input rate 1759000 bits/sec, 350 packets/sec
  5 minute output rate 2122000 bits/sec, 383 packets/sec

this interface is "up/down" since about a day, and happily forwarding
packets.  It does *not* do CDP, which is how we noticed (we monitor
CDP relationships to see topology/device changes).

This is a WS-CBS3020-HPQ, running cbs30x0-ipbase-mz.122-53.SE1, and
the interface config is really trivial:

interface GigabitEthernet0/24
 description Trunk to sw1-int.blau:gi0/24
 switchport trunk allowed vlan 79-82,87-89
 switchport mode trunk
 media-type internal
end

... and no matter what I do (shut/no shut on either end, change 
media-type to external/internal) will bring the port back to up/up.

Shutting down the other end will bring the port to down/down, so it's
still noticing something :-) - but unshutting will bring it back to up/down.

"errdisable autorecovery" didn't help either.

Weird, eh?


Question to the list: has one of you seen this, and found a way to un-stick
the port without rebooting?

gert
-- 
USENET is *not* the non-clickable part of WWW!
                                                           //www.muc.de/~gert/
Gert Doering - Munich, Germany                             gert at greenie.muc.de
fax: +49-89-35655025                        gert at net.informatik.tu-muenchen.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 305 bytes
Desc: not available
URL: <https://puck.nether.net/pipermail/cisco-nsp/attachments/20110202/cfd3ef4c/attachment.pgp>