Tony Lill
2009-Dec-03 16:24 UTC
[Xen-users] 3.4.2 domU networking fails to recover after glitch
A couple of times yesterday, my network hiccuped (something funky on the switch, a couple of other hosts reported the same thing). While the dom0 handled it fine, I could not communicate to any of my domUs. using tcpdump on both eth2 and peth2, I could see them send packets out, but could not see any of the response packets. These are the messages that I get around the hiccup Dec 2 10:52:34 xen kernel: skge peth2: Link is down. Dec 2 10:52:34 xen kernel: eth2: port 1(peth2) entering disabled state Dec 2 10:52:36 xen kernel: skge peth2: Link is up at 1000 Mbps, full duplex, flow control tx and rx Dec 2 10:52:36 xen kernel: eth2: port 1(peth2) entering learning state Dec 2 10:52:36 xen kernel: eth2: topology change detected, propagating Dec 2 10:52:36 xen kernel: eth2: port 1(peth2) entering forwarding state I haven''t had time to do any serious debugging, but I''m hoping someones has seen similar problems. -- Tony Lill, Tony.Lill@AJLC.Waterloo.ON.CA President, A. J. Lill Consultants (519) 650 0660 539 Grand Valley Dr., Cambridge, Ont. N3H 2S2 (519) 241 2461 --------------- http://www.ajlc.waterloo.on.ca/ ---------------- _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
Andrew Lyon
2009-Dec-03 16:56 UTC
Re: [Xen-users] 3.4.2 domU networking fails to recover after glitch
On Thu, Dec 3, 2009 at 4:24 PM, Tony Lill <ajlill@ajlc.waterloo.on.ca> wrote:> A couple of times yesterday, my network hiccuped (something funky on > the switch, a couple of other hosts reported the same thing). While > the dom0 handled it fine, I could not communicate to any of my > domUs. using tcpdump on both eth2 and peth2, I could see them send > packets out, but could not see any of the response packets. > > These are the messages that I get around the hiccup > > Dec 2 10:52:34 xen kernel: skge peth2: Link is down. > Dec 2 10:52:34 xen kernel: eth2: port 1(peth2) entering disabled state > Dec 2 10:52:36 xen kernel: skge peth2: Link is up at 1000 Mbps, full duplex, flow control tx and rx > Dec 2 10:52:36 xen kernel: eth2: port 1(peth2) entering learning state > Dec 2 10:52:36 xen kernel: eth2: topology change detected, propagating > Dec 2 10:52:36 xen kernel: eth2: port 1(peth2) entering forwarding state > > I haven''t had time to do any serious debugging, but I''m hoping > someones has seen similar problems. > > -- > Tony Lill, Tony.Lill@AJLC.Waterloo.ON.CA > President, A. J. Lill Consultants (519) 650 0660 > 539 Grand Valley Dr., Cambridge, Ont. N3H 2S2 (519) 241 2461 > --------------- http://www.ajlc.waterloo.on.ca/ ---------------- > > > > _______________________________________________ > Xen-users mailing list > Xen-users@lists.xensource.com > http://lists.xensource.com/xen-users >Try this: ip link set dev eth2 up or ifconfig eth2 up In my case the problem happened for a different reason, I had run the network-bridge script manually to create a second bridge for eth1, but the physical link (peth1) went down for a while between running the script and starting a domU, that seemed to cause the bridge to get stuck in a down state and required the above command to bring it back up. I''m not sure if this is intended behaviour, I need to read up on what is supposed to happen if a physical interface connected to a bridge goes down.. Andy _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users