Question about OpenVZ and paths

[rugby]

I'm experimenting with a Kerio MailServer KVM and trying to move it to an OpenVZ CentOS container. The question I have is regarding paths and *where* the data is going. If I'm in my OpenVZ container I've given 300GB of space to, I type df and get

simfs / 10%
none /dev


The path for the KMS datastore is /opt/kerio/mailserver

If I migrate to that same path, the container fills up too quickly, but the web admin says I've only used like 5% of my space.

I can put the datastore whereever I need to, I just need to modify a few .conf files within KMS and it's cool. I'm just not sure how an OPenVZ container handles space and *where* I should put the files when I migrate over.

I used to run an ESXi server and now I've got 5 CentOS KVM's on my server and would like to try OPenVZ and see if performance improves for my VM's.

 

[tom]

I'm experimenting with a Kerio MailServer KVM and trying to move it to an OpenVZ CentOS container. The question I have is regarding paths and *where* the data is going. If I'm in my OpenVZ container I've given 300GB of space to, I type df and get

simfs / 10%
none /dev


The path for the KMS datastore is /opt/kerio/mailserver

If I migrate to that same path, the container fills up too quickly, but the web admin says I've only used like 5% of my space.

I can put the datastore whereever I need to, I just need to modify a few .conf files within KMS and it's cool. I'm just not sure how an OPenVZ container handles space and *where* I should put the files when I migrate over.

I used to run an ESXi server and now I've got 5 CentOS KVM's on my server and would like to try OPenVZ and see if performance improves for my VM's.

I do not get your problem. how do you copy the data to your container? just copy it to the path needed (via ssh into the container).

 

[rugby]

Here's what's going on. I set up an OpenVZ VM and gave it 200GB of space. I run df on the machine and see this:

[root@kerio6 ~]# df
Filesystem 1K-blocks Used Available Use% Mounted on
simfs 30540016 7763824 22776192 26% /
none 12337256 4 12337252 1% /dev


The webadmin shows this:

btw, I'm using rsync.

 

[tom]

Here's what's going on. I set up an OpenVZ VM and gave it 200GB of space. I run df on the machine and see this:

[root@kerio6 ~]# df
Filesystem 1K-blocks Used Available Use% Mounted on
simfs 30540016 7763824 22776192 26% /
none 12337256 4 12337252 1% /dev


The webadmin shows this:

btw, I'm using rsync.

yes, 7,6 GB uses (of a total of 200GB) - nothing wrong here.

what version do you run? (pveversion -v)

btw, please use df -h (human readable)

 

[tom]

yes, 7,6 GB uses (of a total of 200GB) - nothing wrong here.

what version do you run? (pveversion -v)

btw, please use df -h (human readable)

found an issue, I will keep you informed as soon as I got it in detail (maximum displayed size with df -h is 137 GB)

 

[rugby]

found an issue, I will keep you informed as soon as I got it in detail (maximum displayed size with df -h is 137 GB)

[root@kerio6 ~]# df -h
Filesystem Size Used Avail Use% Mounted on
simfs 30G 11G 19G 37% /
none 12G 4.0K 12G 1% /dev

This is the OPenVZ machine with 200GB of hard drive space from above.

 

[tom]

found an issue, I will keep you informed as soon as I got it in detail (maximum displayed size with df -h is 137 GB)

ups, I am wrong.

my machine has only 137 GB free space, therefore only 137 GB are available inside. so its working as expected.

 

[tom]

[root@kerio6 ~]# df -h
Filesystem Size Used Avail Use% Mounted on
simfs 30G 11G 19G 37% /
none 12G 4.0K 12G 1% /dev

do you have only 30 GB on the host?

what do you get on the host? I have just 136 available, therefore here is my limit.

proxmox2:~# df -h
Filesystem            Size  Used Avail Use% Mounted on
...
/dev/mapper/pve-data  356G  220G  [COLOR=Red]136G[/COLOR]  62% /var/lib/vz

 

[rugby]

pm01:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/pve/root 16G 998M 14G 7% /
tmpfs 12G 0 12G 0% /lib/init/rw
udev 10M 2.8M 7.3M 28% /dev
tmpfs 12G 0 12G 0% /dev/shm
/dev/mapper/pve-data 36G 18G 18G 50% /var/lib/vz
/dev/sda1 504M 34M 446M 7% /boot


I think that's the problem, I carved out 64GB of my 8TB raid array to install. Is it possible to use the other storage like I do for my KVM's?

 

[tom]

pm01:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/pve/root 16G 998M 14G 7% /
tmpfs 12G 0 12G 0% /lib/init/rw
udev 10M 2.8M 7.3M 28% /dev
tmpfs 12G 0 12G 0% /dev/shm
/dev/mapper/pve-data 36G 18G 18G 50% /var/lib/vz
/dev/sda1 504M 34M 446M 7% /boot


I think that's the problem, I carved out 64GB of my 8TB raid array to install. Is it possible to use the other storage like I do for my KVM's?

no, that's not possible. openvz does not support the flexible storage model. the fastest way would be a reinstallaton with a bigger installation partition.

 

[rugby]

no, that's not possible. openvz does not support the flexible storage model. the fastest way would be a reinstallaton with a bigger installation partition.

Yikes. I've got 5 KVM's running on this box in a datacenter. There's no way I can reinstall anything.

Shoot, one of my CentOS KVM's is having performance issues and I thought this would be a quick fix.

 

[tom]

Yikes. I've got 5 KVM's running on this box in a datacenter. There's no way I can reinstall anything.

Shoot, one of my CentOS KVM's is having performance issues and I thought this would be a quick fix.

if you are familar with and partitioning and LVM you can change also a running system.

 

[rugby]

if you are familar with and partitioning and LVM you can change also a running system.

I think if I change to an e1000 network card that would help this particular VM. I'm going to try that tonight and see if it helps. It's only this one VM, the other 4 are running fine.