Proxmox 4.2 : error with container

[larsislars]

Hello, i have an error with proxmox 4.2
when i create a CT in my ZFS local i have "
mounting container failed
TASK ERROR: cannot open directory //pool2: No such file or directory
"

Or pool2 exist, i have all my VM in my local ZFS.

Can you help me please ?

 

[dcsapak]

could you please post the config of the container in question?

pct config <ID>

 

[larsislars]

could you please post the config of the container in question?

pct config <ID>

I have not ID.
When i create the CT, log create an error and the machine is not create in my pve

 

[dcsapak]

okay, could you please post the content of /etc/pve/storage.cfg and the output of

pvesm status

and maybe the settings with which you try to create the container

 

[DynFi User]

I have the same error.
Fresh install setup in cluster (2 nodes for the time being).

I have deleted the zpool completely and re-created It… no luck still the same error (looks like a bug).

pvesm status

root@proxmonster:/home/gregober# pvesm status
local dir 1 48712192 1013120 47699072 2.58%
local-zfs zfspool 1 47699268 96 47699172 0.50%
mondata zfspool 1 3398564295 191 3398564104 0.50%
montemplate dir 1 3398955904 391808 3398564096 0.51%
newmail_data nfs 1 4184603264 111739520 4072863744 3.17%
tide_vzbackup nfs 1 6729065856 624486016 6104579840 9.78%
​

Zpool status output :

root@proxmonster:/home/gregober# zpool status
pool: monster
state: ONLINE
scan: none requested
config:

NAME STATE READ WRITE CKSUM
monster ONLINE 0 0 0
raidz2-0 ONLINE 0 0 0
sda ONLINE 0 0 0
sdb ONLINE 0 0 0
sdc ONLINE 0 0 0
sdd ONLINE 0 0 0
sde ONLINE 0 0 0
sdg ONLINE 0 0 0
logs
sdf1 ONLINE 0 0 0
cache
sdf2 ONLINE 0 0 0
spares
sdh AVAIL

errors: No known data errors

pool: rpool
state: ONLINE
scan: none requested
config:

NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
mirror-0 ONLINE 0 0 0
sdi2 ONLINE 0 0 0
sdj2 ONLINE 0 0 0

errors: No known data errors​

and zfs list :

root@proxmonster:/home/gregober# zfs list
NAME USED AVAIL REFER MOUNTPOINT
monster 384M 3.17T 192K /md
monster/data 192K 3.17T 192K /md/data
monster/templates 383M 3.17T 383M /md/templates
rpool 8.28G 45.5G 96K /rpool
rpool/ROOT 989M 45.5G 96K /rpool/ROOT
rpool/ROOT/pve-1 989M 45.5G 989M /
rpool/data 96K 45.5G 96K /rpool/data
rpool/swap 7.31G 52.8G 64K -​

Container creation are very basic…
Templates comes from directory "montemplate"

?

 

[DynFi User]

I think I have more or less found the error.

If I create my lxc container by hand, I can create It ok with this command (where I have specified "volume=") :

pct create 117 /md/templates/template/cache/centos-6-default_20160205_amd64.tar.xz \
-description LXC -rootfs volume=/md/data,size=8G -hostname BLnew -memory 1024 -nameserver 192.168.210.25 \
-net0 name=eth0,ip=192.168.210.203/24,bridge=vmbr1 -storage mondata -password

​

Same command without the "-rootfs volume=" will pop the same error :

root@proxmonster:/home/gregober# pct create 117 /md/templates/template/cache/centos-6-default_20160205_amd64.tar.xz \
> -description LXC -rootfs 8 -hostname BLnew -memory 1024 -nameserver 192.168.210.25 \
> -net0 name=eth0,ip=192.168.210.203/24,bridge=vmbr1 -storage mondata -password
Enter password: ********
Retype password: ********
mounting container failed
cannot open directory //monster: No such file or directory​

 

[DynFi User]

I have yet found another reply which is the good and right answer :

Filesystem and mount points shall have the same naming !


So this won't work :

root@proxmonster:/md/data# zfs list
NAME USED AVAIL REFER MOUNTPOINT
monster 384M 3.17T 192K /md
monster/data 240K 3.17T 240K /md/data
monster/templates 383M 3.17T 383M /md/templates
rpool 8.28G 45.5G 96K /rpool
rpool/ROOT 990M 45.5G 96K /rpool/ROOT
rpool/ROOT/pve-1 990M 45.5G 990M /
rpool/data 96K 45.5G 96K /rpool/data
rpool/swap 7.31G 52.8G 64K -​

Where this will perfectly work :

root@proxmonster:~# zfs list
NAME USED AVAIL REFER MOUNTPOINT
monster 1.38M 3.17T 192K /monster
monster/data 192K 3.17T 192K /monster/data
monster/templates 192K 3.17T 192K /monster/templates
rpool 8.28G 45.5G 96K /rpool
rpool/ROOT 990M 45.5G 96K /rpool/ROOT
rpool/ROOT/pve-1 990M 45.5G 990M /
rpool/data 96K 45.5G 96K /rpool/data
rpool/swap 7.31G 52.8G 64K -
​

monster --> /monster
monster/data --> /monster/data
monster/templates --> /monster/templates


This is not really in line with ZFS where you should have independent naming and mount_points


Beside the fact that It took me 1/2 day to understand how It is working, It works

 

[Daeden]

This fixed my issue! Thank you Greg!

 

[larsislars]

Please, can you help me.

root@pve:~# pvesm status
Local_ZFS zfspool 1 1885339648 653582009 1231757639 35.17%
Vauban_ISO dir 1 944198084 645511660 250723904 72.52%
local dir 1 184094804 3625652 180452768 2.47%


root@pve:~# zfs list
NAME USED AVAIL REFER MOUNTPOINT
pool2 623G 1.15T 41.5K /mnt/pool2
pool2/vm-100-disk-1 56.7G 1.20T 4.75G -
pool2/vm-101-disk-1 98.0G 1.23T 11.6G -
pool2/vm-101-disk-2 206G 1.35T 16K -
pool2/vm-103-disk-1 84.1G 1.18T 39.0G -
pool2/vm-103-disk-4 51.6G 1.20T 16K -
pool2/vm-104-disk-1 51.6G 1.15T 50.4G -
pool2/vm-106-disk-1 74.3G 1.15T 67.5G -
pool2/vol3 722M 1.15T 16K -



root@pve:~# zpool status
pool: pool2
state: DEGRADED
status: One or more devices could not be used because the label is missing or
invalid. Sufficient replicas exist for the pool to continue
functioning in a degraded state.
action: Replace the device using 'zpool replace'.
see: http://zfsonlinux.org/msg/ZFS-8000-4J
scan: none requested
config:

NAME STATE READ WRITE CKSUM
pool2 DEGRADED 0 0 0
raidz1-0 DEGRADED 0 0 0
scsi-3600508b1001c39eacef3ee0a8fc4ee26 ONLINE 0 0 0
8469805204689490886 UNAVAIL 0 0 0 was /dev/da3

errors: No known data errors

 

[Rooker]

Filesystem and mount points shall have the same naming !

I think I'm having the same issue
I'm using ZFS and changed to mount-root to "/mnt" (I don't like non-standard folders created in /).

How can I configure Proxmox to be aware of this?

 

[jtl]

I think I'm having the same issue
I'm using ZFS and changed to mount-root to "/mnt" (I don't like non-standard folders created in /).

How can I configure Proxmox to be aware of this?

In the same situation.

 

[remyd1]

Hi,

I have the same problem here with Proxmox 4.4. I created a zpool named "zfs_bigvol" with the same zfs name and same pve storage name/id.

Restoring by specifying the rootfs volume path actually works, as specified by @Greg Bernard .

Thanks,
Rémy

 

[Tekthis]

Just ran into this same issue on 5 - ZFS allows mountpoint name to be different from the poolname, but proxmox assumes the mountpoint is the same as the pool name. Can we have this added to 'things to fix'?

 

[fabian]

Just ran into this same issue on 5 - ZFS allows mountpoint name to be different from the poolname, but proxmox assumes the mountpoint is the same as the pool name. Can we have this added to 'things to fix'?

assuming the mountpoint is the default one saves us a round trip of asking ZFS every time we want to access a container volume.. but I think it should be possible to add this as a (non-default) option, so feel free to file an enhancement request at bugzilla.proxmox.com for the pve-storage component

 

[M4dMike]

Thanks Greg, saved me today.

 

[C_J]

You can use zfs to move the mountpoint, e.g.

zfs set mountpoint=/mainpool mainpool

Verify with

zfs list

I removed the entries from Proxmox's storage menu first and recreated them, out of an abundance of caution.