PLease HELP recover container from cancelled vzmigrate

H

hisaltesse

Well-Known Member
Mar 4, 2009
227
2
58
Hey guys,

I need some urgent help. I ran the command vzmigrate <node> <containerID> in order to migrate a container to a different note.

However because hit CTRL-C to cancel the process so I could do it from a different terminal.

I thought the process would gracefully cancel.. once cancelled, my 60Gb container on the first node disappeared completely.

Is there a way I can recover it? How come vzmigrate would delete a container upon UNsuccessful completion?

Your help would be greatly appreciated.

Thanks.
 
No it was not.. it is a 60GB container.
I ran the command vzmigrate <nodeip> <containerID>
But just a few seconds into it I cancelled the command.

So my logical thought is that the container would remain were it was because vzmigrate was cancelled.
On the contrary it was not copied into the new node and it disappeared from the current node..

When I look at /etc/vz/conf it shows <containerid>.conf.destroyed.

It does not seem like this is an expected behavior. I am trying to figure out how I can recover something.
Shouldn't that container be somewhere in the main node in temp location?
 
hisaltesse said:
It does not seem like this is an expected behavior. I am trying to figure out how I can recover something.
Shouldn't that container be somewhere in the main node in temp location?
Click to expand...

No, there is no temp location.

Anyways, on PVE2 You should use the following command to migrate containers between cluster nodes:

# pvectl migrate <vmid> <target>
 
Thanks for your advice.

However we use proxmox 1.9.

After some digging into the filesystem I found some partial files inside /var/lib/vz/vztmp

I am not sure if these are partial files from a previous failed backup or from the migration I attempted to do.

Do you know what exactly writes in vztmp? Is it backup or migrate or both?
 
hisaltesse said:
When I look at /etc/vz/conf it shows <containerid>.conf.destroyed.
Click to expand...

That file is created at the very end of the vzmigrate script - after a successful rsync. I can't see how that can happen when you abort the script.
 
hotwired007 said:
Have you looked on the second node? btw are you using shared storage?
Click to expand...

No storage is not shared.. but more importantly this does not happen from the interface. It happened from running the vzmigrate command from the terminal (so I could copy 60GB through a private network instead of the internet).
 
You must log in or register to reply here.
Top Bottom