Can not start contaners

LXD
1

Hello
I have a lxd from snap (rev. 13300) , containers can not start after reboot server.

lxc info --show-log m-vw
lxc 20200208164829.378 TRACE commands - commands.c:lxc_cmd:303 - Connection refused - Command “get_state” failed to connect command socket

For storage used btrfs on /dev/sde2.
After few experiments i found a working scheme :

mount /dev/sde2 /var/snap/lxd/common/lxd/storage-pools/btrfs-prod
mount /var/snap/lxd/common/lxd/storage-pools/btrfs-prod/containers /var/snap/lxd/common/lxd/containers/ -o bind

after this, containers starts normaly , but i cant do export.

lxc export  m-vw ./m-vw_06_02_2020.tgz 
Error: Create backup: Backup create: Failed to run: rsync -a -HA --sparse --devices --delete --checksum --numeric-ids --xattrs --bwlimit 0 -vi /var/snap/lxd/common/lxd/storage-pools/btrfs-prod/containers/m-vw/ /var/snap/lxd/common/lxd/backups/lxd_backup_862616164/container: rsync: change_dir "/var/snap/lxd/common/lxd/storage-pools/btrfs-prod/containers/m-vw" failed: No such file or directory (2)
rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1183) [sender=3.1.1]

lxc query /1.0/storage-pools?recursion=1

    {
            "config": {
                    "source": "/dev/sde2"
            },
            "description": "",
            "driver": "btrfs",
            "locations": [
                    "node1",
                    "node2",
                    "node3"
            ],
            "name": "btrfs-prod",
            "status": "Pending",
            "used_by": [
                    "/1.0/containers/m-vw",
                    "/1.0/images/7b58622614fa724290eb15c139501394c63641e81411c13d166825cc8c7fae45",
                    "/1.0/images/f91c8741dcff5597d23d63eec9860285fc3cde52b7a9cfaea88aa4850a9a2658",
                    "/1.0/profiles/btrfs",
                    "/1.0/profiles/btrfs2",
                    "/1.0/profiles/btrfs_prod1"
            ]
    },

How i can export container for transfer to another node or fix this situation ?

2

it seems that you have a bigger problem - you did not create your storage according to

if you have precious work in your container, maybe you could try to save it by using lxc publish (that is create an image from it, that can be used to recreate a new container afterwards).
If it works then zap your cluster attempt and recreate it following the script.