On Fri, Jul 08, 2011 at 10:52:58AM +0100, Richard W.M. Jones wrote: > On Fri, Jul 08, 2011 at 02:55:46PM +0530, Amit Shah wrote: > > On (Thu) 07 Jul 2011 [17:53:51], Richard W.M. Jones wrote: > > > On Thu, Jul 07, 2011 at 09:09:24AM +0100, Richard W.M. Jones wrote: > > > > > > > > On Thu, Jul 07, 2011 at 10:19:22AM +0530, Amit Shah wrote: > > > > > > > > <7>[ 867.432147] PM: Entering mem sleep > > > > > > > > > > The serial console should continue working after resume from S3... can > > > > > you try that? > > > > > > > > Yeah, I'll have to configure a serial console. Will try this > > > > later ... > > > > > > Serial port hangs as well. The last messages on the serial console were: > > > > > > [ 901.147074] PM: early resume of devices complete after 5.725 msecs > > > [ 901.147907] pci 0000:00:01.0: PIIX3: Enabling Passive Release > > > [ 901.148078] ata_piix 0000:00:01.1: setting latency timer to 64 > > > [ 901.148213] uhci_hcd 0000:00:01.2: PCI INT D -> Link[LNKD] -> GSI 10 > > > (level,0 > > > [ 901.148260] uhci_hcd 0000:00:01.2: setting latency timer to 64 > > > [ 901.148423] usb usb1: root hub lost power or was reset > > > [ 901.463151] usb 1-1: reset full speed USB device using uhci_hcd and > > > address 2 > > > [ 901.743467] PM: resume of devices complete after 596.007 msecs > > > [ 901.755059] PM: Finishing wakeup. > > > [ 901.755061] Restarting tasks ... done. > > > > Looks like it came up fine; did you have a tty running on this serial > > line? That should work too. > > Yes, I had an open shell at the time. Hitting return didn't get me > back to the prompt. It looks like the VM / qemu is genuinely locked up. > > I'm a bit confused about this. It appears that sending the shutdown > ACPI signal is causing the VM to suspend *AND* resume (is that right?) > But nothing I'm doing should be causing it to resume, since I'm only > clicking on the SPICE console and hitting the keyboard after the > resume has happened.
Yes, qemu does the resume immediately, not as a result of any keyboard or other external trigger. > > I verified that the resume happens 'spontaneously' by using > virt-dmesg, which uses a non-invasive kernel memory scan. > > Rich. > > -- > Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones > virt-top is 'top' for virtual machines. Tiny program with many > powerful monitoring features, net stats, disk stats, logging, etc. > http://et.redhat.com/~rjones/virt-top > _______________________________________________ > virt mailing list > [email protected] > https://admin.fedoraproject.org/mailman/listinfo/virt _______________________________________________ virt mailing list [email protected] https://admin.fedoraproject.org/mailman/listinfo/virt
