Failed to rename container in startup config and then apply it `copy startup running`

[mattiaswal]

Current Behavior

I had a container (docker://mattiaswal/supertuxkart-dedicated) on RPI4, renamed the container in startup-config, copy startup running no fail, but no change in running config

Expected Behavior

New name in running config and in the system.

Steps To Reproduce

No response

Additional information

No response

[troglobit]

Reproducible

Added a simple nftables contain called before:

admin@infix-00-00-00:/> show log
...
Oct 20 10:03:15 infix-00-00-00 finit[1]: Started container:before[4606]
Oct 20 10:03:15 infix-00-00-00 before[4606]: 
Oct 20 10:03:15 infix-00-00-00 before[4606]: curiOS v25.06.0 (TTY/secure)
Oct 20 10:03:15 infix-00-00-00 before[4606]: 
Oct 20 10:03:15 infix-00-00-00 before[4606]: > SYSTEM INITIATED
Oct 20 10:03:15 infix-00-00-00 before[4606]: > ERROR: ruleset configuration file not found (/etc/nftables.conf)
Oct 20 10:03:15 infix-00-00-00 before[4606]: > SYSTEM MONITORING STANDBY -- No active firewall!

admin@infix-00-00-00:/> show container 
CONTAINER ID  IMAGE                                                COMMAND     CREATED        STATUS        PORTS       NAMES
6be50c89ad92  localhost/curios-nftables-oci-amd64-v25.06.0:latest              7 seconds ago  Up 8 seconds              before

admin@infix-00-00-00:/> copy running-config startup-config 
admin@infix-00-00-00:/> exit
admin@infix-00-00-00:~$ mg /cfg/startup-config.cfg 
# renamed 'before' -> 'after'
admin@infix-00-00-00:~$ 
admin@infix-00-00-00:~$ cli

See the 'help' command for an introduction to the system

admin@infix-00-00-00:/> copy startup-config running-config 
admin@infix-00-00-00:/> show container 
CONTAINER ID  IMAGE                                                COMMAND     CREATED             STATUS             PORTS       NAMES
6be50c89ad92  localhost/curios-nftables-oci-amd64-v25.06.0:latest              About a minute ago  Up About a minute              before

admin@infix-00-00-00:/> reboot
...

After reboot:

admin@infix-00-00-00:/> show container 
CONTAINER ID  IMAGE                                                COMMAND     CREATED         STATUS         PORTS       NAMES
daef165e390c  localhost/curios-nftables-oci-amd64-v25.06.0:latest              41 seconds ago  Up 42 seconds              after

admin@infix-00-00-00:/> show container all 
CONTAINER ID  IMAGE                                                COMMAND     CREATED         STATUS                     PORTS       NAMES
6be50c89ad92  localhost/curios-nftables-oci-amd64-v25.06.0:latest              2 minutes ago   Exited (0) 53 seconds ago              before
daef165e390c  localhost/curios-nftables-oci-amd64-v25.06.0:latest              43 seconds ago  Up 44 seconds                          after

Workaround

So a reboot "fixes" the problem, and since tracking of 'before' is lost, the user can manually clean up by calling

admin@infix-00-00-00:/> container prune 
WARNING! This command removes:
	- all stopped containers
	- all networks not used by at least one container
	- all dangling images
	- all dangling build cache

Are you sure you want to continue? [y/N] y
Deleted Containers
6be50c89ad9294c3e50c85c2e17dccd6349f72f91b60b75a20189748a26ac81d
Total reclaimed space: 4.241kB

Analysis

This issue does not seem to be specific to containers and the recontain branch. To test the theory I edited startup-config again, this time disabling the SSH service. As you can see below, the result was the same.

admin@infix-00-00-00:~$ mg /cfg/startup-config.cfg 
# Set SSH service to "enabled": false
admin@infix-00-00-00:~$ cli
admin@infix-00-00-00:/> copy startup-config running-config 
admin@infix-00-00-00:~$ initctl |grep ssh
3704  sshd             running  [---2345----] OpenSSH daemon
admin@infix-00-00-00:~$ reboot
...
Infix OS — Immutable.Friendly.Secure v25.09.0-48-ge6be413e (hvc0)
infix-00-00-00 login: admin
Password: 
.-------.
|  . .  | Infix OS — Immutable.Friendly.Secure
|-. v .-| https://kernelkit.org
'-'---'-'

admin@infix-00-00-00:~$ initctl |grep ssh
admin@infix-00-00-00:~$ 

Possibly related to #373 and #981

[troglobit]

Found a fix for this, it seems when copying to the running datastore we cannot use sr_copy_config(), instead we must use sr_replace_config().