pct restore: unable to parse config line

[postcd]

# pct restore 101 /vzdump-3810.tgz
unable to parse config line: ▒▒▒▒=

i know this config line is unimportant, how can i proceed pct the restore even there is that bad config line please?

Also in which file this config line is inside that openvz .tgz?

 

[dietmar]

Just to be sure, I wonder if container 101 already exists? If so, please post the content of

# cat /etc/pve/lxc/101.conf

Also in which file this config line is inside that openvz .tgz?

./etc/vzdump/vps.conf

I guess you need to modify that file inside the archive to be able to extract it.

 

[postcd]

101 not existed, this was new server.
> I guess you need to modify that file inside the archive to be able to extract it.

i spent some time trying to extract is and then update it back to archive, but repeatedly ending up with error: "archive contains no configuration file"
gzip -d archive.tgz
tar -xf archive.tar ./etc/vzdump/vps.conf
..modify..
tar -uf archive.tar /etc/vzdump/vps.conf
tar -czf vzdump.tgz archive.tar
pct list;pct restore 105 archive.tgz
-> "archive contains no configuration file"

unsure what i did wrong, maybe paths syntax or i could delete file instead of updating. If anyone know how, please let me know.

 

[dietmar]

The vps.conf need to be the first file inside the archive.

# mkdir data
# cd data
# tar xpf ../archive.tgz
# ... modify...
# tar -czpf ../archive-new.tgz ./etc/vzdump/vps.conf .

 

[postcd]

worked, thx alot

 

[Sebastian's]

I have been trying for few days move our current containers based on openvz CentOS 7 to new LXC Proxmox 5.2. I have backup of all containers in tarball. Each container is in separate tarball backup from /vz/private/VEID

When I tried command :
pct restore NEW-VMID vzdump-file

I end up with this message
ERROR: archive contains no configuration file


Current containers kernel :
uname -r
3.10.0-514.16.1.vz7.30.15

What it might be wrong?

 

[fabian]

you need to either use 'pct create' or pass all the mandatory options to 'pct restore', or re-pack your archives to include configuration files.

 

[Sebastian's]

I have tar archive once again as was suggested above
tar -czpf ../archive-new.tgz ./etc/vzdump/vps.conf .
Now restore pass this point and I end up with different message:
pct restore 1457 archive-new.tgz -storage containers

file does not look like a template archive: /var/lib/vz/dump/archive-new.tgz

Any suggestions would be greatly appreciated. Thx

 

[Sebastian's]

I have attached also part of tarbal tree. Maybe here is something wrong..

root@ns3111642:/var/lib/vz/dump/ tar -xzvf archive-new.tgz
var/lib/vz/dump/vps.conf/
var/lib/vz/dump/vps.conf/media/
var/lib/vz/dump/vps.conf/aquota.group
var/lib/vz/dump/vps.conf/fastboot
var/lib/vz/dump/vps.conf/boot/
var/lib/vz/dump/vps.conf/.cpt_hardlink_dir_a920e4ddc233afddc9fb53d26c392319/
var/lib/vz/dump/vps.conf/etc/
var/lib/vz/dump/vps.conf/etc/lsb-base-logging.sh
var/lib/vz/dump/vps.conf/etc/mtools.conf
var/lib/vz/dump/vps.conf/etc/warnquota.conf
var/lib/vz/dump/vps.conf/etc/bash_completion.d/
var/lib/vz/dump/vps.conf/etc/bash_completion.d/initramfs-tools
var/lib/vz/dump/vps.conf/etc/bash_completion.d/axi-cache
var/lib/vz/dump/vps.conf/etc/bash_completion.d/debconf
var/lib/vz/dump/vps.conf/etc/bash_completion.d/apache2.2-common
var/lib/vz/dump/vps.conf/etc/bash_completion.d/upstart
var/lib/vz/dump/vps.conf/etc/bash_completion.d/insserv
var/lib/vz/dump/vps.conf/etc/bash_completion.d/apt-show-versions
var/lib/vz/dump/vps.conf/etc/sudoers.d/
var/lib/vz/dump/vps.conf/etc/sudoers.d/README
var/lib/vz/dump/vps.conf/etc/quotatab
var/lib/vz/dump/vps.conf/etc/apache2/
var/lib/vz/dump/vps.conf/etc/apache2/sites-available/
var/lib/vz/dump/vps.conf/etc/apache2/sites-available/default
var/lib/vz/dump/vps.conf/etc/apache2/sites-available/default-ssl
var/lib/vz/dump/vps.conf/etc/apache2/httpd.conf
var/lib/vz/dump/vps.conf/etc/apache2/sites-enabled/
var/lib/vz/dump/vps.conf/etc/apache2/sites-enabled/000-default
var/lib/vz/dump/vps.conf/etc/apache2/mods-available/
var/lib/vz/dump/vps.conf/etc/apache2/mods-available/disk_cache.conf
var/lib/vz/dump/vps.conf/etc/apache2/mods-available/alias.conf
var/lib/vz/dump/vps.conf/etc/apache2/mods-available/ssl.load
var/lib/vz/dump/vps.conf/etc/apache2/mods-available/dav_fs.load
var/lib/vz/dump/vps.conf/etc/apache2/mods-available/authnz_ldap.load

 

[fabian]

those paths are not correct.. check an existing backup or template archive to see how it is supposed to look like.

 

[Sebastian's]

Thank you for all suggestions. I have passed now restore a process. Container at the end of process showed up in GUI.
When I start in GUI or CLI I get following messages:

pct restore 8888 8888.tar --rootfs containers:70 -storage containers
extracting archive '/var/lib/vz/dump/8888.tar'
Total bytes read: 56113152000 (53GiB, 173MiB/s)
Detected container architecture: amd64


root@ns3111642:~# pct start 8888
Job for pve-container@8888.service failed because the control process exited with error code.
See "systemctl status pve-container@8888.service" and "journalctl -xe" for details.
command 'systemctl start pve-container@8888' failed: exit code 1
root@ns3111642:~#


root@ns3111642:~# systemctl status pve-container@8888.service
● pve-container@8888.service - PVE LXC Container: 8888
Loaded: loaded (/lib/systemd/system/pve-container@.service; static; vendor preset: enabled)
Active: failed (Result: exit-code) since Sat 2018-12-01 01:21:08 UTC; 7s ago
Docs: man:lxc-start
man:lxc
manct
Process: 25635 ExecStart=/usr/bin/lxc-start -n 8888 (code=exited, status=1/FAILURE)

gru 01 01:21:07 ns3111642 systemd[1]: Starting PVE LXC Container: 8888...
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Control process exited, code=exited status=1
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Killing process 25638 (lxc-start) with signal SIGKILL.
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Killing process 25741 (apparmor_parser) with signal SIG
gru 01 01:21:08 ns3111642 systemd[1]: Failed to start PVE LXC Container: 8888.
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Unit entered failed state.
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Failed with result 'exit-- Subject: pve-container@8888.service -- Defined-By: systemd
-- Support: https://www.debian.org/support
--
-- pve-container@8888.service
--
-- failed.
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Unit entered failed state.
gru 01 01:21:08 ns3111642 pvedaemon[14057]: unable to get PID for CT 8888 (not running?)
gru 01 01:21:08 ns3111642 systemd[1]: pve-container@8888.service: Failed with result 'exit-code'.
gru 01 01:21:08 ns3111642 pct[25633]: command 'systemctl start pve-container@8888' failed: exit code 1
gru 01 01:21:08 ns3111642 pct[25632]:
vzstart:8888:root@pam
gru 01 01:21:08 ns3111642 systemd-timesyncd[1147]: Synchronized to time server 213.251.128.249:123 (ntp.ovh.net).
gru 01 01:21:11 ns3111642 sshd[25842]: Invalid user deploy from 212.224.125.240 port 34395
gru 01 01:21:11 ns3111642 sshd[25842]: input_userauth_request: invalid user deploy [preauth]
gru 01 01:21:11 ns3111642 sshd[25842]: pam_unix(sshd:auth): check pass; user unknown
gru 01 01:21:11 ns3111642 sshd[25842]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser=
gru 01 01:21:13 ns3111642 sshd[25842]: Failed password for invalid user deploy from 212.224.125.240 port 34395 ssh2
gru 01 01:21:13 ns3111642 sshd[25842]: Received disconnect from 212.224.125.240 port 34395:11: Bye Bye [preauth]
gru 01 01:21:13 ns3111642 sshd[25842]: Disconnected from 212.224.125.240 port 34395 [preauth]
gru 01 01:21:42 ns3111642 sshd[26217]: Invalid user o2 from 51.68.140.29 port 58150
gru 01 01:21:42 ns3111642 sshd[26217]: input_userauth_request: invalid user o2 [preauth]
gru 01 01:21:42 ns3111642 sshd[26217]: pam_unix(sshd:auth): check pass; user unknown
gru 01 01:21:42 ns3111642 sshd[26217]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser=
gru 01 01:21:44 ns3111642 sshd[26217]: Failed password for invalid user o2 from 51.68.140.29 port 58150 ssh2
gru 01 01:21:44 ns3111642 sshd[26217]: Received disconnect from 51.68.140.29 port 58150:11: Bye Bye [preauth]
gru 01 01:21:44 ns3111642 sshd[26217]: Disconnected from 51.68.140.29 port 58150 [preauth]
lines 1539-1575/1575 (END)-code'.


/var/lib/vz/dump# cd 8888
/var/lib/vz/dump/8888# cd etc
/var/lib/vz/dump/8888/etc# cd vzdump
/var/lib/vz/dump/8888/etc/vzdump# ls
pct.conf
/var/lib/vz/dump/8888/etc/vzdump# cat pct.conf
arch: amd64
cores: 1
hostname: serv100
memory: 512
net0: name=eth0,bridge=vmbr0,hwaddr=EA:05:03:EB:0A:4E,type=veth
ostype: ubuntu
rootfs: containers:subvol-105-disk-0,size=70G
swap: 512

 

[fabian]

what happens if you start the container in foreground/debugging mode via lxc-start?

 

[Sebastian's]

Here is output of CLI and I attached log file.
Thx

lxc-start -n 8888 -F -l DEBUG -o /tmp/lxc-8888.log
lxc-start: 8888: cgroups/cgfsng.c: create_path_for_hierarchy: 1211 The cgroup "/sys/fs/cgroup/systemd//lxc/8888" already existed
lxc-start: 8888: cgroups/cgfsng.c: cgfsng_create: 1325 Failed to create cgroup "/sys/fs/cgroup/systemd//lxc/8888"
lxc-start: 8888: cgroups/cgfsng.c: create_path_for_hierarchy: 1211 The cgroup "/sys/fs/cgroup/systemd//lxc/8888-1" already existed
lxc-start: 8888: cgroups/cgfsng.c: cgfsng_create: 1325 Failed to create cgroup "/sys/fs/cgroup/systemd//lxc/8888-1"
lxc-start: 8888: cgroups/cgfsng.c: create_path_for_hierarchy: 1211 The cgroup "/sys/fs/cgroup/systemd//lxc/8888-2" already existed
lxc-start: 8888: cgroups/cgfsng.c: cgfsng_create: 1325 Failed to create cgroup "/sys/fs/cgroup/systemd//lxc/8888-2"
lxc-start: 8888: cgroups/cgfsng.c: create_path_for_hierarchy: 1211 The cgroup "/sys/fs/cgroup/systemd//lxc/8888-3" already existed
lxc-start: 8888: cgroups/cgfsng.c: cgfsng_create: 1325 Failed to create cgroup "/sys/fs/cgroup/systemd//lxc/8888-3"
lxc-start: 8888: sync.c: __sync_wait: 59 An error occurred in another process (expected sequence number 7)
lxc-start: 8888: start.c: __lxc_start: 1948 Failed to spawn container "8888"
lxc-start: 8888: tools/lxc_start.c: main: 330 The container failed to start
lxc-start: 8888: tools/lxc_start.c: main: 336 Additional information can be obtained by setting the --logfile and --logpriority options

 

[fabian]

the log says what is wrong - your rootfs does not contain an /sbin/init - most likely your doctored template/backup archive still has wrong paths and init ended up somewhere else.

 

[Ismael Barros²]

I'm sorry to revive and old-ish thread but I also had the problem that was silently solved in pct restore: unable to parse config line and couldn't find the solution until I read the source code of src/PVE/LXC/Create.pm:

    if ($archive =~ /\.tar(\.[^.]+)?$/) {
        if (defined($1)) {
        @compression_opt = $compression_map{$1}
            or die "unrecognized compression format: $1\n";
        }
    } else {
        die "file does not look like a template archive: $archive\n";
    }

pct restore won't accept a .tgz file.

The solution for me was to rename the file so that it ends in .tar.gz instead of .tgz.

 

[grin]

I'm sorry to revive and old-ish thread but I also had the problem that was silently solved in pct restore: unable to parse config line and couldn't find the solution until I read the source code of src/PVE/LXC/Create.pm:
...

pct restore won't accept a .tgz file.

The solution for me was to rename the file so that it ends in .tar.gz instead of .tgz.

Exactly same happened to me. PVE openvz created a tgz while pct cannot handle it, only tar.gz.