Open lcx opened 3 days ago
I've just tested on two separate containers, using both Apple and SMS MFA codes and didn't have a problem.
Can you check your /config folder and see if there are any files that look like your email address and delete them, if there are? I'm suspecting there are some leftovers from a previous run, specifically a .session file.
everything looks fine.
root@diskstation:/volume1/docker/icloudpd/cristian/config# ls -lFa
total 4
drwxr-xr-x+ 1 cristian users 26 Jul 3 13:40 ./
drwxr-xr-x+ 1 cristian users 24 Jul 3 13:08 ../
-rwxr-xr-x 1 root root 1936 Jul 3 13:40 icloudpd.conf*
I just moved the whole config to a Debian Linux with Docker, I have the same results, so I'm definitely doing something wrong here and it's my config or something I'm doing. I will trash my config and start from scratch, the absence of some error is making it hard to know where to start with the debugging. Just to be clear on what I'm doing, this is on a fresh system:
docker compose up
docker exec -ti icloud-cris-icloudpd_cris-1 ash
sync-icloud.sh --Initialise
rechecked the documentation and I'm also missing this message:
ERROR Cookie does not exist."
INFO - Please create your cookie using the --Initialise script command line option."
INFO - Syntax: docker exec -it <container name> sync-icloud.sh --Initialise"
INFO - Example: docker exec -it icloudpd sync-icloud.sh --Initialise"
INFO Restarting in 5 minutes..."
but my config directory has only the mentioned icloudpd.conf
quick update: I tried the docker create version ... without compose and just minimal parameters (only apple_id) and then it worked. So I will remove setting after setting to see when it starts working. It seems like one of the env settings I have added causes this.
found it, as soon as I add this ENV setting it's stuck.
- group=users
Can you set debug_logging=true
in your icloudpd.conf file?
The users:100 group already exists inside the container as it's a default group created by the system. In theory, it should accept this just fine. However, I noticed that one of the error logs (a log type that should always be displayed) is set to be a debug log (which are only shown when the above variable is set. I'm thinking that you may be hitting this error and this is why you see nothing before your container restarts (after 2mins at a guess).
Looking at your yaml though, this shouldn't be happening. I should know more once you add debug logging though.
Edit: Also, I found out the other day that this is actually deprecated now. Can safely remove it from your yaml.
version: "3.8"
2024-07-03 13:36:51 INFO Cookie path: /config/cristianlivadarunet
A shot in the dark but you could try creating the folder named cristianlivadarunet in the config folder with proper ownership and not world readable (750).
A shot in the dark but you could try creating the folder named cristianlivadarunet in the config folder with proper ownership and not world readable (750).
Minor correction in that /config/cristianlivadarunet
is a file, not directory. I think you should be able to get away with 600 permissions too, as the file should always be witten/read by the local user in the container, who should be the owner.
I'm not being prompted for password after --init and after a while the container is just stopped and is being restarted.
Here is my compose file:
anything I'm doing wrong here? I can't see any error or other issues. Also I'm trying to run this on a synology.