[Xen-users] Weird arp problem

[Manuel SUBREDU <manuel.subredu@xxxxxxxxx>]

 Hi,

 we have a very strange problem. We have a HP C7000 blade enclosure with
ProLiant BL460c G6 blades within. Each blade has 8 cores, 32GB of memory
and 2 network cards, connected trough two ethernet passtrough modules
into a Cisco Catalyst 3750G series. On the blades, we have OpenSuSE 11.2
x86_64 with stock xen kernel.
 The problem: random, when we start or stop a virtual machine, the dom0
network connection hangs. Dom0 doesn't respond to ping or anything else.
After some time (about a minute), the network connection works again,
without us doing something special.
 We manage to connect trough the blade management module on the machine
when the network connection does not work and found out that if we
remove the gateway arp entry from the arp table (using ip neigh del),
the network connection starts working right away.
 Did anyone found this problem too ? What can be the cause of this ?

-- 
=====================================
Manuel SUBREDU
Network Engineer at AARNIEC/RoEduNet
RoCSIRT Coordinator
mobile: +40 (758) 808 052
phone:  +40 (232) 201 007
fax:    +40 (232) 201 200
e-mail: manuel.subredu@xxxxxxxxx
=====================================

signature.asc
Description: OpenPGP digital signature

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users