lxd: alpine containers fails to start
This is a very similar issue to #15290 (closed), but I can reproduce it on my system on the latest Alpine edge, even though #15290 (closed) was fixed two weeks ago. Neither the existing containers nor newly created containers with default settings no longer start after my first upgrade since end of August / beginning of September (everything worked flawlessly before that upgrade):
~ $ doas lxc launch images:alpine/edge simple-lxd-test
doas (inga@localhost) password:
Creating simple-lxd-test
Starting simple-lxd-test
Error: Failed to run: /usr/sbin/lxd forkstart simple-lxd-test /var/lib/lxd/containers /var/log/lxd/simple-lxd-test/lxc.conf: exit status 1
Try `lxc info --show-log local:simple-lxd-test` for more info
~ $ doas lxc info --show-log simple-lxd-test
doas (inga@localhost) password:
Name: simple-lxd-test
Status: STOPPED
Type: container
Architecture: x86_64
Created: 2023/10/13 12:23 CEST
Last Used: 2023/10/13 12:23 CEST
Log:
lxc simple-lxd-test 20231013102319.635 WARN cgfsng - ../src/lxc/cgroups/cgfsng.c:fchowmodat:1619 - No such file or directory - Failed to fchownat(18, memory.oom.group, 1000000000, 0, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW )
lxc simple-lxd-test 20231013102319.635 WARN cgfsng - ../src/lxc/cgroups/cgfsng.c:fchowmodat:1619 - No such file or directory - Failed to fchownat(18, memory.reclaim, 1000000000, 0, AT_EMPTY_PATH | AT_SYMLINK_NOFOLLOW )
lxc simple-lxd-test 20231013102319.731 ERROR cgfsng - ../src/lxc/cgroups/cgfsng.c:cgfsng_mount:2139 - No such file or directory - Failed to create cgroup at_mnt 24()
lxc simple-lxd-test 20231013102319.732 ERROR conf - ../src/lxc/conf.c:lxc_mount_auto_mounts:852 - No such file or directory - Failed to mount "/sys/fs/cgroup"
lxc simple-lxd-test 20231013102319.732 ERROR conf - ../src/lxc/conf.c:lxc_setup:4433 - Failed to setup remaining automatic mounts
lxc simple-lxd-test 20231013102319.732 ERROR start - ../src/lxc/start.c:do_start:1272 - Failed to setup container "simple-lxd-test"
lxc simple-lxd-test 20231013102319.732 ERROR sync - ../src/lxc/sync.c:sync_wait:34 - An error occurred in another process (expected sequence number 4)
lxc simple-lxd-test 20231013102319.735 WARN network - ../src/lxc/network.c:lxc_delete_network_priv:3631 - Failed to rename interface with index 0 from "eth0" to its initial name "vetha92d9236"
lxc simple-lxd-test 20231013102319.735 ERROR start - ../src/lxc/start.c:__lxc_start:2107 - Failed to spawn container "simple-lxd-test"
lxc simple-lxd-test 20231013102319.735 WARN start - ../src/lxc/start.c:lxc_abort:1036 - No such process - Failed to send SIGKILL via pidfd 19 for process 5506
lxc simple-lxd-test 20231013102319.735 ERROR lxccontainer - ../src/lxc/lxccontainer.c:wait_on_daemonized_start:878 - Received container state "ABORTING" instead of "RUNNING"
~ $
A workaround suggested in #15290 (closed) still works; if I do doas umount /sys/fs/cgroup/systemd
prior to the attempt to lxc launch
/ lxc start
, the containers are started successfully.