OpenVZ Forum


Home » General » Support » Failed ot lookup path /proc, vzctl restore not working (vzctl restore generates error)
Failed ot lookup path /proc, vzctl restore not working [message #51595] Mon, 11 August 2014 14:39 Go to next message
JimL is currently offline  JimL
Messages: 116
Registered: February 2007
Senior Member
From: *hsd1.ga.comcast.net
Note the error in the subject line isn't my typo. "To" is misspelled in the error printed by the system.

[root@chinaberry jwl]# lst /tmp
total 63556
-rw------- 1 root  root  54986624 Aug 11 10:19 204.vz
[root@chinaberry jwl]# vzctl --version
vzctl version 4.7.2


root@chinaberry jwl]# vzmigrate 192.168.2.55 204
Locked CT 
Starting migration of CT 204 to 192.168.2.55
Preparing remote node
Initializing remote quota
vzquota : (warning) Quota file exists, it will be overwritten
Syncing private
Stopping container
Syncing 2nd level quota
Starting container
Error: Failed to start container
Starting container...
Container is mounted
Adding IP address(es): 192.168.2.4
Setting CPU units: 1000
Configure veth devices: veth204.0 
Container start in progress...
vzquota : (error) Quota is not running for id 204
[root@chinaberry jwl]# vzctl chkpnt 204 --dumpfile /tmp/204.vz
Setting up checkpoint...
	suspend...
	dump...
	kill...
Checkpointing completed successfully
Container is unmounted
[root@chinaberry jwl]# lst /tmp
total 63556
-rw------- 1 root  root  54986624 Aug 11 10:19 204.vz



Copied it to another system:

Linux centos 2.6.32-042stab072.10 #1 SMP Wed Jan 16 18:54:05 MSK 2013 x86_64 x86_64 x86_64 GNU/Linux
[root@centos root]# vzctl --version
vzctl version 4.7.2



[root@centos root]# vzctl restore 204 --dumpfile /root/204.vz
Restoring container ...
Initializing quota ...
Container is mounted
	undump...
Adding IP address(es): 192.168.2.4
Setting CPU units: 1000
Configure veth devices: veth204.0 
Restore error, undump failed: No such file or directory
Error: Failed ot lookup path '/proc'
Error: Failed to restore mount point @3528 dev 'proc', type 'proc', path '/proc'
Error: rst_namespace: -2
Container restore failed
Container is unmounted
[root@centos root]# ls /vz/private/204
[root@centos root]# ls /vz/root/204



Note, I was following the instructions at https://openvz.org/Checkpointing_and_live_migration. When moving from one system to another you have to manually transfer the config file and make the /vz/root/<VID> and /vz/private/<VID> directories. I think it should be mentioned on that page.

I also attempted a migration as
[root@chinaberry jwl]# vzmigrate 192.168.2.55 204
Locked CT 
Starting migration of CT 204 to 192.168.2.55
Preparing remote node
Initializing remote quota
vzquota : (warning) Quota file exists, it will be overwritten
Syncing private
Stopping container
Syncing 2nd level quota
Starting container
Error: Failed to start container
Starting container...
Container is mounted
Adding IP address(es): 192.168.2.4
Setting CPU units: 1000
Configure veth devices: veth204.0 
Container start in progress...
vzquota : (error) Quota is not running for id 204


But there was no trace of 204 on the destination system. No config, no root or public directories.

Thanks for any help

Jim.
Re: Failed ot lookup path /proc, vzctl restore not working [message #51596 is a reply to message #51595] Tue, 12 August 2014 03:29 Go to previous message
devonblzx is currently offline  devonblzx
Messages: 127
Registered: December 2006
Senior Member
From: 162.220.217*
You don't have any data in the /vz/private/204 directory so the server won't be able to load. Checkpointing only saves the memory to the dump, not the hard drive.

It looks like you are just trying to migrate so the easy way is just to use vzmigrate with the --live option which will use checkpointing along with the hard drive transfer.

As for the migration failing, enable verbose so you can see more details as to why it is failing but I would recommend doing this before migrating:



  • Remove /vz/private/204 and /vz/root/204
  • Remove /etc/vz/conf/204.conf


Then see if it works. I have seen the migration fail quite often when the server data already exists on the destination node.

If a live (online) migration fails, then it is most likely due to processor differences or kernel differences and you'll just have to do an offline migration which will require a couple minutes of downtime depending on how big the virtual server is. The key to having live migration work is using a similar kernel and similar processor.


http://static.openvz.org/userbars/openvz-user-2.png
ByteOnSite President

[Updated on: Tue, 12 August 2014 03:40]

Report message to a moderator

Previous Topic: First Openvz Installation
Next Topic: disk size becomes weird after set diskspace many times
Goto Forum:
  


Current Time: Thu Dec 05 18:02:50 GMT 2019