Lxc copy to remote of different (higher) version

[Simon_Wibberley]

It is possible to copy to an LXD remote of a high version?

Many thanks!

[tomp (Thomas Parrott)]

Yes that should be fine.

[Simon_Wibberley]

Thanks Thomas,

It appears I get “Error: Get https://1xxx.xxx.xxx.xxx:8443: remote error: tls: protocol version not supported”

Going from 3.6 → 5.1

Any ideas?

Many thanks,

Simon

[tomp (Thomas Parrott)]

See

Tls remote connection problem

You can set LXD_INSECURE_TLS=true in the environment of the LXD 5.0 system. Easiest way to do this would be systemctl edit snap.lxd.daemon and then add a section like: [Service] Environment=LXD_INSECURE_TLS=true To get it applied, you could try systemctl reload snap.lxd.daemon which may do the trick. If it doesn’t, a full system reboot is likely the easiest alternative.

[Simon_Wibberley]

Thanks again Thomas!

That has cleared the TLS issue, however it now hangs on the copy command.

local:
[root@shl1 ~]# lxc copy git-http lxd3:sftp

remote:

[root@shl2 ~]# lxc list
+------+---------+------+------+-----------+-----------+
| NAME |  STATE  | IPV4 | IPV6 |   TYPE    | SNAPSHOTS |
+------+---------+------+------+-----------+-----------+
| sftp | STOPPED |      |      | CONTAINER | 0         |
+------+---------+------+------+-----------+-----------+

There doesn’t appear to be any network activity.

Could you advise on diagnosing further?

Kind regards,

Simon

[tomp (Thomas Parrott)]

Try doing lxc export <instance> <file name>.tar.gz and then transfer the file manually to target and run lxc import <file name>.tar.gz on the target host.

[Simon_Wibberley]

Hmm, I’m getting:

[root@shl1 lxd]# lxc export git-http git-http.tgz -vvv
Error: Create backup: Backup storage: Could not create snapshot LV named snapshot--05236fd3--6b5f--4499--b074--7ae20cfb9d88

The reason I’m trying to backup containers is because I’m in the below situation. My sense is the above error might be related?..

Do you this I should reboot the host to recover state?

Many thanks,

Simon

Nobody nogroup after cancelled publish

I cancelled an lxc publish as the partition it was using was about to run out of space, then after starting the container again all uid have been lost with nobody nogroup and services do no start: [root@shl1 lxd]# lxc exec omeka-s bash bash: /root/.bashrc: Permission denied any clues on how to fix uid mappings? Thanks in advance For some reason: /var/snap/lxd/common/mntns/var/snap/lxd/common/lxd/storage-pools/default/containers/NAME/rootfs/ is empty. config: image.architecture: amd64 …

[tomp (Thomas Parrott)]

Yes try a reboot

[Simon_Wibberley]

The above error persists after a reboot

Thanks for your helps so far - any advice on diagnosing further?

Kind regards,

Simon

[tomp (Thomas Parrott)]

Is the container running when you run lxc export?

[Simon_Wibberley]

Yes it was.

[tomp (Thomas Parrott)]

Try stopping it and doing export as that might skip the snapshot.

[Simon_Wibberley]

Alas, it appears to want to snapshot when STOPPED too…

[tomp (Thomas Parrott)]

Can you create a new storage pool of dir type and then copy the instance to it locally, then try exporting it from there.

[Simon_Wibberley]

[root@shl1 lxd-export]~ lxc export git-http-bak git-http-bak.tgz
Backup exported successfully!           

Fab, thanks so much!!

[Simon_Wibberley]

Nearly there! I can export the smaller ones, but even when exporting to a different partition lxd wants to consume space on the root partition which is nearly full. Is there some way to only use the target partition?

Many thanks,

Simon

[Simon_Wibberley]

Just found this!

Root space usage during export

My LXD server is an Ubuntu server (18.04), LXD is installed with snap package and my storage is ZFS. When I try to export a large container (> 90 GB) to a NFS share (to backup it) the operation fail because my host root space is filled up (19 GB free). Is it normal ? What can I do ? Can we move “all” LXD directories to ZFS storage because with this configuration I can’t make large images ? Thanks

[tomp (Thomas Parrott)]

See Lxc export behavior - #2 by tomp

[Simon_Wibberley]

Thanks Thomas,

I’m getting

[root@shl1 sw206]# lxc config set storage.backups_volume pool2/storage
Error: cannot set 'storage.backups_volume' to 'pool2/storage': unknown key

I’m on LXD 3.6 - was this config on a different key back then?

Many thanks,

Simon

[tomp (Thomas Parrott)]

Ah you cannot upgrade from 3.x to 5.x (only from 4.x), as 3.x is very old.

You could try upgrading the 3.x host to 4.x and then exporting and reimporting.