Nico Prenzel
2007-Nov-28 07:23 UTC
[Xen-devel] How to debug XEN, in case where DomU/Dom0 gets unresponsive?
Hi all, I would like to know how to debug xen. To give some more informations, I would like to explain why I''d like to do this. I''ve compiled xen from source (3.1.x/unstable) and installed it on Debain etch. Furthermore I''ve installed 2 DomU, one HVM (Win2k3) and one EFW (Endian Firewall, RH-based 2.6.9-55 kernel). Both DomU''s are running fine for about 15 minutes. But that differs from time to time. If something happens (like file transfer or http traffic) then all DomUs and the Dom0 get unresponsive. I think the "hang" wouldn''t be xen related, if only the DomUs would crash and the Dom0 would reachable. But with the described problem the Dom0 also gets unresponsive. Furthermore the DomUs weren''t crashed, because the console of EFW/"xm top" details this. Anymore I can''t see any logs at Dom0 that the bridges went down. A restart of the Dom0 fixes the problem as it appears again the next time. So, would there be any way to debug this, I think xen-related, problem? Greetings. NicoP. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Mark Williamson
2007-Dec-03 03:47 UTC
Re: [Xen-devel] How to debug XEN, in case where DomU/Dom0 gets unresponsive?
> I would like to know how to debug xen. > To give some more informations, I would like to explain why I''d like to do > this. > I''ve compiled xen from source (3.1.x/unstable) and installed it on Debain > etch. Furthermore I''ve installed 2 DomU, one HVM (Win2k3) and one EFW > (Endian Firewall, RH-based 2.6.9-55 kernel). > Both DomU''s are running fine for about 15 minutes. But that differs from > time to time. If something happens (like file transfer or http traffic) > then all DomUs and the Dom0 get unresponsive. > I think the "hang" wouldn''t be xen related, if only the DomUs would crash > and the Dom0 would reachable. But with the described problem the Dom0 also > gets unresponsive. Furthermore the DomUs weren''t crashed, because the > console of EFW/"xm top" details this. Anymore I can''t see any logs at Dom0 > that the bridges went down. A restart of the Dom0 fixes the problem as it > appears again the next time. > So, would there be any way to debug this, I think xen-related, problem?Do you have any direct physical access to the machine? Or maybe a remotely accessible serial line? That would make fact-finding easier if (for instance) the dom0 network connection has died for some reason but the system is actually still up. Otherwise, I guess you could check the Xend logs for anything suspicious and the system logs for any kernel oopses? From my reading of the code for xenconsoled you should be able to run "xenconsoled -l hv" in order to get it to log messages that come out of "xm dmesg" into /var/log/xen/hypervisor.log which you could also inspect after any crash... I can''t find a more friendly way of activating that logging. Cheers, Mark -- Dave: Just a question. What use is a unicyle with no seat? And no pedals! Mark: To answer a question with a question: What use is a skateboard? Dave: Skateboards have wheels. Mark: My wheel has a wheel! _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Apparently Analagous Threads
- xenconsoled needs restart after long dom0 uptime otherwise domU console output hangs and domU doesn''t come up
- "''Invalid kernel'', ''xc_dom_find_loader: no loader found\n" with new vanilla 2.6.23-rc8
- XEN 4.0.2-rc2 - using OCAML xenstored -- xend not starting?
- build error with current (changeset 16475) xen-unstable tree
- Cannot start xen domUs anymore, domUs hang on kernel startup, happens after a long dom0 uptime