You can tail sched.log, to see if the scheduler puts the VM in migrate
mode... but I don't think this can help you a lot.
I just resoved a similar issue on my installation with xen: originally I
didn't set a fixed memory for Dom0 and, then, OpenNebula monitor didn't
recognize the real amount of free ram on the host: in fact OpenNebula
thinked the host was full of ram because Linux Bufferization and, then,
didn't migrate the VM.
You can check this via onehost list command: if you see a small amount of
RAM, less than VM RAM, this is the problem.
Hope this can help you,
Alberto Zuin
http://www.azns.it
----- Original Message -----
From: "Thomas Knauth" <thomas.kna...@gmx.de>
To: <users@lists.opennebula.org>
Sent: Monday, June 27, 2011 11:23 PM
Subject: [one-users] VM stuck in "migration" state - migration never
issued(Xen)
Hello list,
I am experiencing very long delays (>15min) between issuing $ onevm
livemigrate ... and the actual migration taking place. Which parts of
ONE are taking part in the decision when to migrate?
The log file of the VM says
Mon Jun 27 17:33:57 2011 [LCM][I]: New VM state is MIGRATE
but nothing happens. Migrations do work in general. Sometimes. When it
does not work, the
command "xm -l ..." (using Xen) is never executed on the host the VM
is running on (as can be confirmed by looking into /var/log/auth.log).
Which parts of ONE are involved when I issue the "onevm livemigrate"
command from the console? Where can I get more debug info to
troubleshoot the problem?
Thanks in advance,
Thomas.
_______________________________________________
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
_______________________________________________
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
