windows network freeze

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

windows network freeze


we ran hosts on xen 4.1 (on debian) for many years, about ten hosts (all linux debian) and one windows (windows 2008 R2).

After upgrade to 4.8 we start having issues with networking on windows host, after a while (once a few hours , once it was ok whole weekend): Windows say that have no internet connection, but "cable is connected", I can't ping from windows to dom0 or back from dom0 to windows, I can see in brctl show same state as it was working:

(in vif*-emu is state UNKOWN, in vif* state DOWN)

34: vif30.0-emu: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master xenbr0 state UNKNOWN group default qlen 1000
      link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff
      inet6 fe80::fcff:ffff:feff:ffff/64 scope link
         valid_lft forever preferred_lft forever
35: vif30.0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master xenbr0 state DOWN group default qlen 32
      link/ether fe:ff:ff:ff:ff:ff brd ff:ff:ff:ff:ff:ff

After reboot of this windows host, network start working again.  All other hosts (linux debian) is working week without problem.

With upgrade to 4.8 we migrated data from LVM to ZFS, It looks like this issue happened when sending zfs snapshots to other hosts (or when is bigger load of system during week days), but it could be wrong thought and zfs don't have any fluence on this.
Config of this host in attachment.

Please can you point me where could be problem?

Thanks, Dusan

Xen-users mailing list
[hidden email]

win2k8.cfg (808 bytes) Download Attachment