Red Hat Bugzilla – Bug 224004
Live migration to localhost leaves guest hung
Last modified: 2010-10-22 04:12:25 EDT
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:220.127.116.11) Gecko/20061221 Fedora/18.104.22.168-1.fc5 Firefox/22.214.171.124
Description of problem:
After migrating a live guest to localhost, the guest appears to be hung.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. xm create -c DOMNAME
2. xm migrate -l DOMNAME localhost
The guest appears to hang.
xm console DOMNAME fails with message xenconsole: Could not read tty from store: No such file or directory
Guest doesn't reply to ping.
xm shutdown does nothing.
Guest doesn't hang. xm console works. Guest replies to ping. xm shutdown works.
xenstore-ls /local/domain/$(xm domid DOMNAME)/device shows all frontend devices in state 1 (XenbusStateInitialising). That's the initial state written by xend. Frontend devices go through intermediate states into state 4 (XenbusStateConnected) when they initialize. Looks like they didn't (re-)initialize at all.
Found on RHEL 5 snapshot 7 updated with kernel-xen-2.6.18-2.el5. Also reproduced on FC-6 with xen-3.0.3-3.fc6 and kernel-xen-2.6.19-1.2895.fc6.
xm save / restore works.
Setting rhel-5.1.0 to '?', because this is a blocker for bug 212662 which also
has rhel-5.1.0 set.
This request was evaluated by Red Hat Product Management for inclusion in a Red
Hat Enterprise Linux maintenance release. Product Management has requested
further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed
products. This request is not yet committed for inclusion in an Update
It is currently unclear whether migration to localhost works with our upstream
base version, or even with the upstream head version.
This bug no longer blocks bug 212662, so the reason why flag rhel‑5.1.0 was
requested (comment#1) no longer exists. I recommend to push this to 5.2.
Markus - Please retest and update.
Retested, same behavior.
Steps to Reproduce:
1. Edit /etc/xen/xend-config.sxp to have:
Default values for the other xend-relocation-* parameters should be fine,
leave them unchaged.
2. Start guest with virt-manager or virsh start.
3. xm migrate -l DOMNAME localhost
virsh list shows guest as paused.
Created attachment 258291 [details]
/var/log/xen/xend.log since reboot
Created attachment 258301 [details]
xenstore-ls after boot, before virsh start
Created attachment 258311 [details]
xenstore-ls after virsh start, before xm-migrate
Created attachment 258321 [details]
xenstore-ls after xm-migrate
This bug could be caused by our fix for bug 245960, which effectively reverted
xen-unstable cset 12576. Need to test that.
Might also have to make the disk shareable.
Unreverting cset 12576 and making the disk shareable don't make a difference.
Withholding QE ack until we sort out a path forward. Appears from comment 14
migrating to localhost isn't really a real-world event and the potential fix
might be very destabilizing.
Cannot fix for RHEL 5.2.
Development Management has reviewed and declined this request. You may appeal
this decision by reopening this request.
the RHEL5.2 release notes will be dropped to translation on April 15, 2008, at
which point no further additions or revisions will be entertained.
a mockup of the RHEL5.2 release notes can be viewed at the following link:
please use the aforementioned link to verify if your bugzilla is already in the
release notes (if it needs to be). each item in the release notes contains a
link to its original bug; as such, you can search through the release notes by