Proxmox 9.1.5 breaks LXC mount points?

B

bluepr0

Well-Known Member
Mar 1, 2019
79
12
48
70
Hi!

I just updated to the latest version of Proxmox (9.1.5), and I've noticed that the LXCs with mount points (in my case, NFS from my NAS) are not starting. If I remove the mount point, it starts correctly. See attached images.

SCR-20260202-myjc.png

if I remove it
SCR-20260202-mynr.png

Media is an NFS share from my NAS on another hardware, using the storage option in Proxmox
1770040272699.png

Debug logs
Bash: 
lxc-start 108 20260202143201.433 INFO     confile - ../src/lxc/confile.c:set_config_idmaps:2295 - Read uid map: type u nsid 0 hostid 100000 range 65536
lxc-start 108 20260202143201.433 INFO     confile - ../src/lxc/confile.c:set_config_idmaps:2295 - Read uid map: type g nsid 0 hostid 100000 range 65536
lxc-start 108 20260202143201.433 INFO     lsm - ../src/lxc/lsm/lsm.c:lsm_init_static:38 - Initialized LSM security driver AppArmor
lxc-start 108 20260202143201.433 INFO     utils - ../src/lxc/utils.c:run_script_argv:587 - Executing script "/usr/share/lxc/hooks/lxc-pve-prestart-hook" for container "108", config section "lxc"
lxc-start 108 20260202143201.812 DEBUG    utils - ../src/lxc/utils.c:run_buffer:560 - Script exec /usr/share/lxc/hooks/lxc-pve-prestart-hook 108 lxc pre-start produced output: failed to propagate uid and gid to mountpoint: Operation not permitted

lxc-start 108 20260202143201.827 ERROR    utils - ../src/lxc/utils.c:run_buffer:571 - Script exited with status 1
lxc-start 108 20260202143201.827 ERROR    start - ../src/lxc/start.c:lxc_init:845 - Failed to run lxc.hook.pre-start for container "108"
lxc-start 108 20260202143201.827 ERROR    start - ../src/lxc/start.c:__lxc_start:2046 - Failed to initialize container "108"
lxc-start 108 20260202143201.828 INFO     utils - ../src/lxc/utils.c:run_script_argv:587 - Executing script "/usr/share/lxcfs/lxc.reboot.hook" for container "108", config section "lxc"
lxc-start 108 20260202143202.330 INFO     utils - ../src/lxc/utils.c:run_script_argv:587 - Executing script "/usr/share/lxc/hooks/lxc-pve-poststop-hook" for container "108", config section "lxc"
lxc-start 108 20260202143202.693 DEBUG    utils - ../src/lxc/utils.c:run_buffer:560 - Script exec /usr/share/lxc/hooks/lxc-pve-poststop-hook 108 lxc post-stop produced output: umount: /var/lib/lxc/.pve-staged-mounts/mp1: not mounted.

lxc-start 108 20260202143202.693 DEBUG    utils - ../src/lxc/utils.c:run_buffer:560 - Script exec /usr/share/lxc/hooks/lxc-pve-poststop-hook 108 lxc post-stop produced output: command 'umount -- /var/lib/lxc/.pve-staged-mounts/mp1' failed: exit code 32

lxc-start 108 20260202143202.733 ERROR    lxc_start - ../src/lxc/tools/lxc_start.c:lxc_start_main:307 - The container failed to start
lxc-start 108 20260202143202.733 ERROR    lxc_start - ../src/lxc/tools/lxc_start.c:lxc_start_main:312 - Additional information can be obtained by setting the --logfile and --logpriority options



Thanks a lot!
 
Last edited:
  • Like
Reactions: altano, meyergru and goxxy
yep, i have the same problem, proxmox 9.1.5 lxc with mountpoints, same error, if i reboot this lxc, working if already running.
run_buffer: 571 Script exited with status 1
lxc_init: 845 Failed to run lxc.hook.pre-start for container "126"
__lxc_start: 2046 Failed to initialize container "126"
TASK ERROR: startup for container '126' failed
Click to expand...
 
That version of pve-container is running? You can see that if you run pveversion -v on the CLI or click on the "Package Versions" button located at the top left of the NODE → Summary panel.

Does it only happen only with bindmounts to a network share or to any generic local directory? Because the latter seems to be working fine after a very quick test with pve-container: 6.1.0.
 
aaron said:
That version of pve-container is running? You can see that if you run pveversion -v on the CLI or click on the "Package Versions" button located at the top left of the NODE → Summary panel.

Does it only happen only with bindmounts to a network share or to any generic local directory? Because the latter seems to be working fine after a very quick test with pve-container: 6.1.0.
Click to expand...
It's pve-container: 6.1.0

With local directory seems to work fine!
 
Last edited:
  • Like
Reactions: goxxy
Having the same issue but my lxc are unprivileged.
Code: 
lxc-start: 137: ../src/lxc/utils.c: run_buffer: 571 Script exited with status 1
lxc-start: 137: ../src/lxc/start.c: lxc_init: 845 Failed to run lxc.hook.pre-start for container "137"
lxc-start: 137: ../src/lxc/start.c: __lxc_start: 2046 Failed to initialize container "137"
lxc-start: 137: ../src/lxc/tools/lxc_start.c: lxc_start_main: 307 The container failed to start
lxc-start: 137: ../src/lxc/tools/lxc_start.c: lxc_start_main: 312 Additional information can be obtained by setting the --logfile and --logpriority options
 
Thank you for this thread — it was extremely helpful.


I ran into exactly the same issue after upgrading to Proxmox VE 9.1.5: LXC mount points (especially bind mounts on NFS shares) suddenly stopped working properly. At first I assumed something was wrong with my own setup, and I spent almost a full day debugging fstab, systemd mount ordering, NFS exports, permissions, and container configs.


Only after reading this discussion did I realize the problem was introduced by the upgrade itself and not by my system. This post confirmed what I was seeing.


The suggested workaround of downgrading/pinning pve-container was very helpful and immediately resolved the issue for me.


Thanks a lot to everyone who contributed and shared solutions — it saved me a lot of time in the end.
 
Thanks a lot !
This downgrade made my day !
I restored my lxc backup for nothing. I lost 3 days of activities, but not a big deal and nothing not recoverable.
 
aaron said:
It looks like you run into the bug reported here: https://bugzilla.proxmox.com/show_bug.cgi?id=7271
Feel free to chime in there!
To temporarily downgrade pve-container you can run
Code: 
apt install pve-container=6.0.18
Click to expand...
This also fixed it for me. The same issue as everyone else.
Got offered a container update, accepted and then couldn't start any containers with host-based mountpoints.
Thank you for the suggestion, Aaron.
 
  • Like
Reactions: altano
If you check the bugtracker, there is a patch available as of today to be code reviewed: https://bugzilla.proxmox.com/show_bug.cgi?id=7271#c6

Once reviewed and applied, it will be part of one of the next versions of the pve-container package. Ideally check the bugtracker for any updates and the changelog for any updates to the pve-container package which can be viewed easily if you select the package in the Update panel in the web ui.
 
  • Like
Reactions: jf2021, Skye0, Johannes S and 2 others
FYI this also affects ceph bind mounts. (I think it's the same underlying issue so I won't open a separate thread/report.)

Relevant config (readonly mount point):

Code: 
mp0: /mnt/pve/cephfs-hdd/dump,mp=/cephfs-hdd-backups,backup=0,ro=1

Starting container:

Code: 
# pct start 105 --debug
run_buffer: 571 Script exited with status 30
lxc_init: 845 Failed to run lxc.hook.pre-start for container "105"
__lxc_start: 2046 Failed to initialize container "105"
0 hostid 100000 range 65536
INFO     lsm - ../src/lxc/lsm/lsm.c:lsm_init_static:38 - Initialized LSM security driver AppArmor
INFO     utils - ../src/lxc/utils.c:run_script_argv:587 - Executing script "/usr/share/lxc/hooks/lxc-pve-prestart-hook" for container "105", config section "lxc"
DEBUG    utils - ../src/lxc/utils.c:run_buffer:560 - Script exec /usr/share/lxc/hooks/lxc-pve-prestart-hook 105 lxc pre-start produced output: failed to propagate uid and gid to mountpoint: Read-only file system

ERROR    utils - ../src/lxc/utils.c:run_buffer:571 - Script exited with status 30
ERROR    start - ../src/lxc/start.c:lxc_init:845 - Failed to run lxc.hook.pre-start for container "105"
ERROR    start - ../src/lxc/start.c:__lxc_start:2046 - Failed to initialize container "105"
INFO     utils - ../src/lxc/utils.c:run_script_argv:587 - Executing script "/usr/share/lxc/hooks/lxc-pve-poststop-hook" for container "105", config section "lxc"
startup for container '105' failed

Fails on 6.0.19 and 6.1; works on 6.0.18
 
Last edited:
You must log in or register to reply here.
Top Bottom
Back