From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeremy Fitzhardinge Subject: Re: Re: No shutdown of domU: xenbus_dev_shutdown: device/console/0: Initialising != Connected, skipping Date: Fri, 19 Nov 2010 09:07:43 -0800 Message-ID: <4CE6AEDF.9040700@goop.org> References: <1341501057.20101117132828@eikelenboom.it> <1945713192.20101117212227@eikelenboom.it> <4CE44FDB.7000508@goop.org> <1266442973.20101118205715@eikelenboom.it> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1266442973.20101118205715@eikelenboom.it> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Sander Eikelenboom Cc: Ian Campbell , Keir Fraser , "Xen-devel@lists.xensource.com" , Ian Jackson List-Id: xen-devel@lists.xenproject.org On 11/18/2010 11:57 AM, Sander Eikelenboom wrote: > Hi Jeremy/Keir, > > I have tried to add the following lines: > if (!strcmp(type, "console")) > return 0; > > But it seems to be a red herring .. although the "the xenbus_dev_shutdown: device/console/0: Initialising != Connected, skipping" warning is gone. > The PV domU guest still doesn't get completely halted: > - It stays in a running state > - It uses 100% cpu according to xentop > - I can still connect to it's console > > So probably something is stuck in a waiting loop, which doesn't sleep in between and consumes 100% cpu. Hm, I generally see clean shutdowns of my multi-VCPU domains. Can you do # gdbsx -a 9999 (arch-size is 32 or 64) then $ gdb vmlinux (gdb) tar rem :9999 (gdb) thread apply all bt to see what's going on? (You may need to enable CONFIG_DEBUG_INFO to get useful info if you haven't already.) J > Last lines of domU's console(with some additional printk's in xenbus_dev_shutdown): > > Cannot access the Hardware Clock via any known method. > Use the --debug option to see the details of our search for an access method. > Stopping enhanced syslogd: rsyslogd. > Asking all remaining processes to terminate...done. > All processes ended within 2 seconds....done. > Deconfiguring network interfaces...done. > Cleaning up ifupdown.... > Deactivating swap...done. > Unmounting local filesystems...done. > Will now halt. > [ 46.643035] md: stopping all md devices. > [ 47.643320] xenbus_dev_shutdown: trying shutdown of device/vif/0: Connected > [ 47.716448] xenbus_dev_shutdown: result of shutdown of device/vif/0: Closed > [ 47.716461] xenbus_dev_shutdown: trying shutdown of device/vbd/51714: Connected > [ 47.772293] xenbus_dev_shutdown: result of shutdown of device/vbd/51714: Closed > [ 47.772306] xenbus_dev_shutdown: trying shutdown of device/vbd/51713: Connected > [ 47.829384] xenbus_dev_shutdown: result of shutdown of device/vbd/51713: Closed > [ 47.829415] System halted. > > > > it's a domU using a 2.6.37-rc2 kernel, file: based disk access, nothing very special. > > the only difference that triggers it is: > vcpus=1 works fine .. > vcpus>1 symptoms above .. > > I have also added xend.log, with first a boot and shutdown with vpcus=1 and then a boot and shutdown with vpcus=4 > I left the domain(with 4vcpus) running with 100% cpu for 10 minutes, but doesn't seem to hit any timeout, and keeps running. > On 20:40 it ends up stuck, on 20:51 i'm shooting the domain off with xm destroy. > > > Any pointers for adding extra debug info ? > > -- > Sander > > > > > Wednesday, November 17, 2010, 10:57:47 PM, you wrote: > >> On 11/17/2010 12:22 PM, Sander Eikelenboom wrote: >>> Ah yes, 2.6.18 does contain the following line in xenbus_probe.c and 2.6.37-rc2 not: >>> >>> - if (!strcmp(type, "console")) >>> - return 0; >>> It seems to be changed, >>> >>> >>> There seems to have been a patch http://lists.xensource.com/archives/html/xen-devel/2009-11/msg01072.html but it seems it's not applied to xenbus in linux upstream. >> And does this patch fix things for you? >> But I have to say that's pretty gross. Is it really the right thing to do? >> J >> >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xensource.com >> http://lists.xensource.com/xen-devel