backup failed with exit code 11

[huky]

I upgraded a pve cluster from v5 to v6 and added a PBS to it.
Now I had backup 3 small lxc (<10G for disk usage), but meet error when big size disk lxc.

the output:
INFO: starting new backup job: vzdump 114 --mode snapshot --node xnode006 --compress zstd --remove 0 --storage pbsXPFnfs
INFO: Starting Backup of VM 114 (lxc)
INFO: Backup started at 2021-06-19 13:44:24
INFO: status = running
INFO: CT Name: fmDB-vlan104.103
INFO: including mount point rootfs ('/') in backup
INFO: mode failure - some volumes do not support snapshots
INFO: trying 'suspend' mode instead
INFO: backup mode: suspend
INFO: ionice priority: 7
INFO: CT Name: fmDB-vlan104.103
INFO: including mount point rootfs ('/') in backup
INFO: starting first sync /proc/5654/root/ to /var/tmp/vzdumptmp12721_114
ERROR: Backup of VM 114 failed - command 'rsync --stats -h -X -A --numeric-ids -aH --delete --no-whole-file --sparse --one-file-system --relative '--exclude=/tmp/?*' '--exclude=/var/tmp/?*' '--exclude=/var/run/?*.pid' /proc/5654/root//./ /var/tmp/vzdumptmp12721_114' failed: exit code 11
INFO: Failed at 2021-06-19 14:07:39
INFO: Backup job finished with errors
TASK ERROR: job errors

the lxc config:
# pct config 114
arch: amd64
cores: 8
hostname: fmDB-vlan104.103
memory: 32768
net0: name=eth0,bridge=vmbr0,gw=172.17.104.254,hwaddr=A2:64:2E:C7:7A:0D,ip=172.17.104.103/24,tag=104,type=veth
ostype: centos
rootfs: VMs:vm-114-disk-1,size=1T
swap: 512

Is there size limit for vm/lxc?

 

[dcsapak]

is there enought tmp space for the whole container in /var/tmp?

check with 'df -h /var/tmp'
if not, set a different tmpdir in /etc/vzdump.conf

 

[huky]

is there enought tmp space for the whole container in /var/tmp?

check with 'df -h /var/tmp'
if not, set a different tmpdir in /etc/vzdump.conf

thanks. you are right.

The same lxc could be backup via NFS，and the diff between PBS and NFS is tmpdir:
NFS sync tmp to NFS but PBS sync tmp to local.

could PBS sync to PBS? I have a 3T lxc, there is not enough local space.

 

[dcsapak]

there is no builtin way to do that, but you can configure any directory in /etc/vzdump.conf (also e.g. a fileshare)
but i would advise against this, since you transfer your container multiple times over the network...
instead, i'd try to do a snapshot backup (works on e.g. zfs/ceph/lvmthin) or a stop backup. both modes do not need to transfer to a tmpdir