Lustre discuss - May 2010 - Ethernet Bonding stop working on lustre server

Hi all,

I have a lustre environment with 4 server and 2 filesystem.
Server 1: mgs - mds:fs01 - failover of mds:fs02
Server 2: mds:fs02 - failover of mgs - failover of mds:fs01
Server 3: oss:fs01 - failover of oss:fs02
Server 4: oss:fs02 - failover of oss:fs01

Each server has 4 ethernet card.
eth0 and eth1 are connected to an iscsi shared storage with multipath.
eth2 and eth3 are aggregate in bond0, in active-backup mode, that 
exports lustre to clients.

After some hours bond0 stop working(no ping from net to server, no ping 
from server to net) until I check the status of bonding ("cat 
/proc/net/bonding/bond0") and I shutdown the active interface ("ifdown
eth2" for example),so backup interface go up and network starts to 
working again.
At this point I''ve tried to reactivate first interface and to
deactivate
the active interface, in this way network works properly.

My kernel version is 2.6.18-164.11.1.el5_lustre.1.8.2 and my lustre 
version is 1.8.2.

Any idea?

Thanks in advance.

Griitings.


-- 
Gianluca Tresoldi
***SysAdmin***
***Demon''s Trainer***
Tuttogratis Italia Spa
E-mail: gianluca.tresoldi at tuttogratis.com
http://www.tuttogratis.it
Tel Centralino 02-57313101
Tel Diretto 02-57313136
<http://www.gnu.org>Be open...
************ Confidentiality Notice & Disclaimer *************
This message, together with any attachments, is for the confidential
and exclusive use of the addressee(s). If you receive it in error,
please delete the message and its attachments from your system
immediately and notify us by return e-mail.
Do not disclose, copy, circulate or use any information contained in
this e-mail.
*
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.lustre.org/pipermail/lustre-discuss/attachments/20100505/03fff034/attachment.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: linux40.jpg
Type: image/jpeg
Size: 1399 bytes
Desc: not available
Url :
http://lists.lustre.org/pipermail/lustre-discuss/attachments/20100505/03fff034/attachment.jpg