[Xen-bugs] [Bug 1350] New: VIF Failure under heavy network load between domUs and dom0

[bugzilla-daemon@xxxxxxxxxxxxxxxxxxx]

http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=1350

           Summary: VIF Failure under heavy network load between domUs and
                    dom0
           Product: Xen
           Version: 3.0 (general)
          Platform: x86
        OS/Version: Linux-2.6
            Status: NEW
          Severity: critical
          Priority: P1
         Component: Hypervisor
        AssignedTo: xen-bugs@xxxxxxxxxxxxxxxxxxx
        ReportedBy: jsimonds@xxxxxxxxx


VIF bridge fails under load.  This bug should be easily reproduced using nfs
traffice from domU to dom0.  I have the dom0 setup as the nfs server serving
out several directories to the domUs.  Any large nfs transfer from dom0 to domU
or domU to dom0 will immediately cause failure of all the VIFs on all domUs on
the server.  Seems heavy traffic from domU to domU may not cause the failure. 
The only resolution is to reboot the dom0.  There is no noticable entries in
the logs of either the dom0 or domU regarding this failure.  All domUs are
still accessable from the dom0 via "xm console" but all network traffic fails
from dom0 to domUs and all domUs are isolated from each other and the internet.
This occured in both 3.2.0 and continues after upgrade to 3.3.0.  I am running
xen-source 2.6.21 for the dom0 and the domUs.  dom0 and domUs are latest
gentoo.  Following the mailing list, others are experiencing this same issue
both with gentoo and other linux distributions.  Others have reported failure
of the VIF bridge with as little as 3Mbit/sec traffic between dom0 and domU. 

Link to thread in the xen-users mailing list:
http://lists.xensource.com/archives/html/xen-users/2008-09/msg00011.html

There has been some speculation as to whether this may be related to only
gigabit nics.  I have this problem with Intel Pro/1000 nics, but others have
reported the problem on various other NICs (Realtek and NetXtreme to name a
couple) in an attempt to rule out driver/NIC issues.

from lspci:
0d:00.0 Ethernet controller: Intel Corporation 82573E Gigabit Ethernet
Controller (Copper) (rev 03)
0e:00.0 Ethernet controller: Intel Corporation 82573L Gigabit Ethernet
Controller

from syslog:
Sep 10 18:58:15 xen Intel(R) PRO/1000 Network Driver - version 7.3.20-k2
Sep 10 18:58:15 xen e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network
Connection
Sep 10 18:58:15 xen e1000: eth1: e1000_probe: Intel(R) PRO/1000 Network
Connection


I will provide any further information that I can upon request.  
Thank you.


-- 
Configure bugmail: 
http://bugzilla.xensource.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

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