Bug 453491

That error message basically means something went funky when kicking off the qemu command line. Do you still see this or was it a one off occasion? If you can reproduce it, can you attach /var/log/libvirt/qemu/{guestname}.log after the error message appears? Thanks

I'm still seeing it on a current F9 system.

Unable to complete install '<class 'libvirt.libvirtError'> virDomainCreateLinux() failed internal error Timed out while reading monitor startup output
Traceback (most recent call last):
  File "/usr/share/virt-manager/virtManager/create.py", line 651, in do_install
    dom = guest.start_install(False, meter = meter)
  File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line 845, in start_install
    return self._do_install(consolecb, meter)
  File "/usr/lib/python2.5/site-packages/virtinst/Guest.py", line 866, in _do_install
    self.domain = self.conn.createLinux(install_xml, 0)
  File "/usr/lib64/python2.5/site-packages/libvirt.py", line 841, in createLinux
    if ret is None:raise libvirtError('virDomainCreateLinux() failed', conn=self)
libvirtError: virDomainCreateLinux() failed internal error Timed out while reading monitor startup output
'

After that, it keeps installing anyway, even after issuing the error dialog.

Here's the contents of that log file:
----

/usr/bin/qemu-kvm -S -M pc -m 512 -smp 1 -name v1 -monitor pty -no-reboot -boot d -drive file=/var/lib/libvirt/images/v1-1.img,if=ide,index=0 -drive file=/dev/sr0,if=ide,media=cdrom,index=2 -net nic,macaddr=00:16:3e:5b:f8:35,vlan=0 -net tap,fd=14,script=,vlan=0,ifname=vnet0 -serial pty -parallel none -usb -vnc 127.0.0.1:0 -k en-us 
char device redirected to /dev/pts/13
char device redirected to /dev/pts/14
iESC[KESC[DinESC[KESC[DESC[DinfESC[KESC[DESC[DESC[DinfoESC[KESC[DESC[DESC[DESC[Dinfo ESC[KESC[DESC[DESC[DESC[DESC[Dinfo cESC[KESC[DESC[DESC[DESC[DESC[DESC[Dinfo cpESC[KESC[DESC[DESC[DESC[DESC[DESC[DESC[Dinfo cpuESC[KESC[DESC[DESC[DESC[DESC[DESC[DESC[DESC[Dinfo cpusESC[K
* CPU #0: pc=0x00000000000ffff0 thread_id=16021^McESC[KESC[DcoESC[KESC[DESC[DconESC[KESC[DESC[DESC[DcontESC[K
----

Btw, why is this full of non-human readable data?

The unreadable output is the result of reading data from the qemu monitor: we should try and sanitize it at the libvirt level.

I'm not sure though what is causing the timeout, there doesn't seem to be anything wrong with the bootup. Other people have reported similar issues and my guess is that it's load dependent: when you create the vm do you specify a new disk so that virt-manager allocates it?

Maybe the fix is just set the timeout from 3 seconds to 10: a true timeout shouldn't be a common occurence, and if the qemu command errors it will exit immediately anyways so we won't hit the timeout issue.

In this case, I believe I created a new disk during the installation, probably whatever the default is.

*** Bug 331511 has been marked as a duplicate of this bug. ***

*** Bug 458734 has been marked as a duplicate of this bug. ***

*** Bug 472099 has been marked as a duplicate of this bug. ***

*** Bug 473158 has been marked as a duplicate of this bug. ***

Okay, I'm pretty sure I figured this out. The reason this is timing out is because we are installing off the cd drive, and the qemu monitor is unresponsive while the media is spinning up.

I think the only thing to do is just lengthen the timeout value at the libvirt level, which I'm going to test now. Re-assigning to libvirt.

Okay, a fix was committed upstream:

http://git.et.redhat.com/?p=libvirt.git;a=commit;h=be1c939a7533e18eb58711f947ea46dcd44473d5

This is in libvirt 0.5.1 which should be in F9 and F10 updates-testing (if not now, then soon). Moving this to ON_QA.

Fixed in 0.5.1, so closing.