[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] Nested HVM in xen-unstable tests

osstest service owner writes ("[xen-unstable test] 64750: regressions - FAIL"):
> flight 64750 xen-unstable real [real]
> http://logs.test-lab.xenproject.org/osstest/logs/64750/
>  test-amd64-amd64-qemuu-nested 16 debian-hvm-install/l1/l2 fail baseline 
> untested

It appears that this new test case is failing in xen-unstable.  The
logs are here:


If you would like to avoid this feature regressing, you probably want
to figure out what is wrong and send patches to fix it.  In the
meantime, the test will continue to fail in osstest but this will
not block pushes and will not impede anyone else's work.

The primary failure symptom detected by osstest is visible here:


Near the end:

  ssh: connect to host port 22: No route to host

Searching for from the top of that log shows this:

  2015-11-19 16:24:15 Z guest l1.guest.osstest: 5a:36:0e:ee:00:15

So, at some point during the L2 install, the L1 has fallen off the
network.  "No route to host" almost certainly means it has stopped
responding to ARP requests.

The L1's `serial console' log (actually, the log captured by the L0 on
the L1' emulated serial port) is here:


That shows

  (XEN) *** Serial input -> Xen (type 'CTRL-x' three times to switch input to 

which is a result of ts-logs-capture sending it the debug keys (which
you can see here


  2015-11-19 16:27:33 Z spawning ssh -o StrictHostKeyChecking=no -o 
BatchMode=yes -o ConnectTimeout=100 -o ServerAliveInterval=100 -o 
PasswordAuthentication=no -o ChallengeResponseAuthentication=no -o 
root@xxxxxxxxxxxxx 'cat 
  2015-11-19 16:27:33 Z xenuse sending request for input to Xen

So the L1 hasn't completely crashed.

During the log capture, ossstest correctly diagnoses that the L1 is
unreachable.  osstest would then normally `power cycle' the `host',
but in the version of osstest used for this test, that functionality
is broken.  This was fixed in my recent set of patches which have
passed the osstest push gate, and we will soon start to see test
reports which are using the fixed version.

You can see a similar failure with better logs here:


Let me know if I can do more to help point you to the right bits of
the logs, etc.  Good luck.


Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.