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

[Xen-devel] Domains go into a strange dying / shutdown state


  • To: xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>
  • From: "Richard W.M. Jones" <rjones@xxxxxxxxxx>
  • Date: Sat, 19 May 2007 13:20:30 +0100
  • Delivery-date: Sat, 19 May 2007 05:18:05 -0700
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>

After doing some stress testing which basically involves starting up and shutting down FC6 paravirt guests running on Xen 3.1.0, I have domains which fall into a strange state where they cannot be shutdown or destroyed.

I modified libvirt so that it prints out the return from the hypervisor call getdomaininfo. The results are below (all domains other than 0 are in this state).

domain = 0
flags = 32
tot_pages = 501888
max_pages = 4294967295
shared_info_frame = 464
cpu_time = 164613775580514
nr_online_vcpus = 4
max_vcpu_id = 3
ssidref = 0
handle = 0x618d94

domain = 656
flags = 196613
tot_pages = 472
max_pages = 65536
shared_info_frame = 4023
cpu_time = 255195882
nr_online_vcpus = 1
max_vcpu_id = 0
ssidref = 0
handle = 0x618ddc

domain = 659
flags = 1
tot_pages = 472
max_pages = 65536
shared_info_frame = 4035
cpu_time = 4036443852
nr_online_vcpus = 1
max_vcpu_id = 0
ssidref = 0
handle = 0x618e24

domain = 866
flags = 1
tot_pages = 472
max_pages = 65536
shared_info_frame = 3970
cpu_time = 3612084955
nr_online_vcpus = 1
max_vcpu_id = 0
ssidref = 0
handle = 0x618e6c

domain = 1210
flags = 1
tot_pages = 472
max_pages = 65536
shared_info_frame = 3998
cpu_time = 1178432030
nr_online_vcpus = 1
max_vcpu_id = 0
ssidref = 0
handle = 0x618eb4

For example:

# /usr/sbin/xm destroy 1210
# /usr/sbin/xm shutdown 1210
Error: Domain '1210' does not exist.
# /usr/sbin/xm destroy 1210

(But if I rerun the HV list command above, I still see the domain 1210).

Some points to note:

* cpu_time doesn't increase between listings.
* The previous time this happened, I had flags = 5 (is_dying && is_shut_down) for all the broken domains. * This problem is readily reproducible if I leave the tests running overnight.

Is there anything I can do to help debug this further?

Rich.

--
Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/
Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod
Street, Windsor, Berkshire, SL4 1TE, United Kingdom.  Registered in
England and Wales under Company Registration No. 03798903

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel

 


Rackspace

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