Hi Carsten,
I started the VM image with kvm and I saw the the problem was that the
VM was not halted correctly (I used onevm shutdown <vm_id>). Therefore
the VM gave error during the boot process. So, to avoid the previous
problem, I logged into the VM and I executed halt. After that I used
the "onevm shutdown <vm_id> ". In this way, the new image was created fine.
Nevertheless, it is interesting what you said about the networks,
because I think that I am having this problem in an VM with Ubuntu
server. I copied the file vmcontext.sh in the /etc/init.d and in
/etc/rc2.d/ but when I start the VM it doesn't get an IP address. I have
just googled and I found that removing the file
"/etc/udev/rules.d/70-/persistent/-net.rules" could solve the problem.
However, if it works and you are using a persistent image, you should
remove this file each time that you deploy such image. Anyway, I will
try this to see what happens.
Thanks,
Javi
El 21/10/2010 18:20, [email protected] escribió:
Hi Javi,
My first guess would be that the network interface in the VM has problems. You
can confirm this by starting the VM with VNC enabled. If it comes up alright
log in through VNC and check the network interface connection.
Hard to guess what went wrong, but one thing you might want to look at: Some Linux
distros (and I don't know if ttylinux is one of them) keep track of what network cards
they have seen (identifying them by their MAC address). The first time a network card is
found, it is bound to a network device, e.g. eth0. If Linux then finds a network card
with a different MAC address it assigns it to a different device, e.g. eth1. It does so
even if the first card is not present (as it interprets this scenario as "a second
card has been added to the machine and the first one is currently (temporarily)
disabled"; which with real network cards that don't change their MAC address is the
most likely scenario). Now, this does not work well with OpenNebula as the VM is likely
to get a different MAC address for each launch; this will be assigned to a new network
device and your /etc/network/interfaces file probably only includes a configuration for
eth0 (which does not exist). If this turns out to be the case, there is a setting in
Linux to prevent it from remembering cards. I forgot what exactly it is, but it should be
easy to Google.
Carsten
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Javier Diaz
Sent: Thursday, 21 October 2010 3:04
To: [email protected]
Subject: [one-users] onevm saveas problem
Hi,
I'm using OpenNebula 2.0 and I'm trying to save a VM with the "onevm
saveas" command. The VM is the ttylinux provided by you as example and
the only modification that I've made is create a file. The problem is
that when I start the saved VM (with the same parameters except the
Image name) I can't access it using ssh. If I execute "onevm show
<id>", Its status is "runn" and it has an private IP address assigned.
Any idea what is happening?
Thanks you in advance.
Regards,
Javi
--
+-----------------------------------------------------+
Dr. Javier Diaz
Post Doctoral Fellow
Pervasive Technology Institute
Indiana University
2719 East Tenth Street, Bloomington, IN 47408, USA
e-mail: [email protected]
web: http://qcycar-uclm.esi.uclm.es/jdiaz
+-----------------------------------------------------+
_______________________________________________
Users mailing list
[email protected]
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
