Strange. Before upgrading to RC1 I had no problems with docker for a couple weeks. I am not sure what to do since all my apps are not running now
I just posted an issue on github
Oct 02 17:27:14 truenas dockerd[8149]: failed to start daemon: Error initializing network controller: error creating default "bridge" network: all predefined address pools have been fully subnetted
Oct 02 17:27:14 truenas systemd[1]: docker.service: Main process exited, code=exited, status=1/FAILURE
How many apps do you have?
I think around 12 apps
I’m not super familiar with the ins and outs of docker, but the log looks like it’s exhausted the default address pools. You might be able to modify the address pool sizes from 27 to 26 in /etc/docker/daemon.json to test but I’m not sure if middleware will play nice, or if it will help.
This is probably going to be one to raise a Jira ticket for so you can do some troubleshooting with someone at iX, though worth noting the logs in your debug look like a different issue to @rambro1stbud’s so a fix for one will probably not work for the other.
1 Like
Thank you for looking into this for me. Did I make the issues in the right place?
Edit: I posted a ticket on Jira. I see there are 2 other tickets about the same problem
Edit: I see 4 reports of similiar problems. Might be a good idea to avoid RC1 until this issues is resolved.
Edit: Just got off a remote sension with ix he was able to fix the problem and I am all good to go!
1 Like
From the WebUI, you can raise a Jira issue here:
You will need to sign up if you don’t already have an account, but this would be the proper place to report bugs like this.
Edit: Ah, didn’t see your edit 
Hopefully this is an easy fix.
1 Like
That’s invalid syntax. You need (at least) an opening services:
At least I did.
And your indentation is wrong.
Never fear, I got that figured.
1 Like
For what it’s worth:
(backup/test system upgraded from 24.04.2.2 to 24.10-RC1)
Around 12 apps, but only 3 were running. After update to RC1, only 2 apps show.
I knew better as to not to believe the migration was going to be perfect, so I plan to redo the apps under Docker, but hoping it can help the IX Team get the migration working better,
After update, Dashy shows an error and the rest of the apps (most not running, are gone).
cat /var/log/app_migrations.log
[2024/10/02 23:24:03] (DEBUG) app_migrations.migrate():225 - Migration details for ‘system-update–2024-10-03_03:11:48’ backup on ‘LOCAL_NVME_1_1b’ pool
[2024/10/02 23:24:03] (DEBUG) app_migrations.migrate():231 - ‘dashy’ app failed to migrate successfully: ‘App with same name is already installed’
[2024/10/02 23:24:03] (DEBUG) app_migrations.migrate():231 - ‘nextcloud’ app failed to migrate successfully: ‘Failed to migrate config: Failed to migrate config: Traceback (most recent call last):\n File “/mnt/.ix-apps/truenas_catalog/trains/stable/nextcloud/1.3.14/migrations/migrate_from_kubernetes”, line 95, in \n print(yaml.dump(migrate(yaml.safe_load(f.read()))))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File “/mnt/.ix-apps/truenas_catalog/trains/stable/nextcloud/1.3.14/migrations/migrate_from_kubernetes”, line 21, in migrate\n “admin_user”: config[“ncConfig”][“adminUser”],\n ~~~~~~^^^^^^^^^^^^\nKeyError: 'ncConfig'\n’
====
Dashy Web page:
It looks like something’s gone wrong…
This is likely caused by the app source not being found at the current domain
If you need additional support, check the browser console then raise a ticket
Okay, it must have been a combination of changing the DNS server and this terminal command:
midclt call -job docker.update '{"address_pools": [{"base": "172.17.0.0/12", "size": 24}]}'
Several people on the JIRA board are reporting this as the fix. I did type this command, but it didn’t work initially until I switched over to Google name servers around a hour later. Your mileage may vary.
Just upgraded from 24.04.2.2 to 24.10-RC1.
I had around 20 apps, half of them running.
After the upgrade the Apps section was not configured. I’ve set the correct volume but I still don’t see any app.
I tried to run the command but it didn’t help.
midclt call -job docker.update ‘{“address_pools”: [{“base”: “172.17.0.0/12”, “size”: 24}]}’
/var/log/app_migrations.log is empty
Am I missing a step to make the migration of the apps?
Did you restart after making that change?
Yes, I tried rebooting 3 times
In that case please open a bug ticket and attach your debug so we can investigate.
We’ve had a few reports of apps not showing up due to a browser cache issue, we recommend clearing the browser cache after upgrade.
Can you rerun the migration and see if that helps:
midclt call -job k8s_to_docker.migrate tank
Replace tank with the name of the pool where your apps are installed.
I hit the exact same issue when I first installed the BETA and had to rerun migration as it had tried running before there was proper network connectivity.
Thanks, indeed a very good hint!
I also changed DNS to 8.8.8.8 and 8.8.4.4
Some apps were migrated while other no. Funny thing, some apps didn’t even tried to get migrated.
Some errors:
This app has not been migrated, however it throws an error for a dataset that already exists
[2024/10/03 17:37:23] (DEBUG) app_migrations.migrate():231 - ‘stable-diffusion-webui-cuda’ app failed to migrate successfully: “Failed to clone and promote ix-volumes: [EFAULT] Failed to clone snapshot: cannot create ‘nvme_volume/ix-apps/app_mounts/stable-diffusion-webui-cuda/workspace’: dataset already exists”
I have no clue for this error:
[2024/10/03 17:37:23] (DEBUG) app_migrations.migrate():231 - ‘tika’ app failed to migrate successfully: ‘Failed to migrate config: Failed to migrate config: Traceback (most recent call last):\n File “/mnt/.ix-apps/truenas_catalog/trains/stable/ix-app/1.0.6/migrations/migrate_from_kubernetes”, line 142, in \n print(yaml.dump(migrate(yaml.safe_load(f.read()))))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File “/mnt/.ix-apps/truenas_catalog/trains/stable/ix-app/1.0.6/migrations/migrate_from_kubernetes”, line 128, in migrate\n if config[“enableUIPortal”]\n ~~~~~~^^^^^^^^^^^^^^^^^^\nKeyError: 'enableUIPortal'\n’
[2024/10/03 17:37:23] (DEBUG) app_migrations.migrate():231 - ‘opencanary’ app failed to migrate successfully: ‘Failed to migrate config: Failed to migrate config: Traceback (most recent call last):\n File “/mnt/.ix-apps/truenas_catalog/trains/stable/ix-app/1.0.6/migrations/migrate_from_kubernetes”, line 142, in \n print(yaml.dump(migrate(yaml.safe_load(f.read()))))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File “/mnt/.ix-apps/truenas_catalog/trains/stable/ix-app/1.0.6/migrations/migrate_from_kubernetes”, line 128, in migrate\n if config[“enableUIPortal”]\n ~~~~~~^^^^^^^^^^^^^^^^^^\nKeyError: 'enableUIPortal'\n’
I had some Chia dockers that I customized. All those dockers were not migrated and not mentioned in the logs. My customization was a custom entrypoint script and the data being stored in a specific dataset on another volume.
Amazing, I’ve been waiting for this for two years 
Migration went great!
2 Likes
Couple of issues I noticed in regards to custom Compose apps:
- big one: no way to edit your app after it was created!!! it just says “Error detected reading App” in the interface
- no way of specifying ENV vars, it’s common for apps to require a .env file, and no, it doesn’t seem
env_file:on all containers works properly (expanding vars in mounts seems broken, plus it’s annoying to copy-paste it to every container section); there could be an input section for ENV file contents under Compose section if nothing else - no way of specifying “base dir” for Compose, often they expect us to link files by relative paths there. That’s not a huge one as this can be adjusted, but still
- minor: after adding new app, it doesn’t appear in the Apps screen before refreshing the page or changing the tab in the dashboard to something else and back to Apps
EDIT: env_file: won’t work due to this: Docker-Compose Environment-Variables blank string - Stack Overflow
Basically env_file sets the envs INSIDE of the container, while default .env adds them for the purposes of variable substitution in the Compose files…