Due to the scope of this update, you may need to reinstall the integration and clear your browser cache for everything to work smoothly. There is a high chance that your existing entities will not be migrated across and any automations will be affected.
I read that on the release notes. Before i installed this update i cleared the intgration, entities and history before updating it so it should have been as a fresh install.
I was just checking because i dont have a lot of time for a proper bug report. I will try another fresh install and do a github issue if it exists still.
1 Like
I have noticed with chrome or edge browser entities goes missing until I clear the cache of HA related stuff and it shows up again.
You can also enable debug logging in the integration, reload it and view HA Settings → System → Logs for any errors or warnings. Filter by unraid or docker or vm
Lots of info in debug logs.
@ruaandeysel
I just got notified of the update (2024.12.27) to this integration, and I saw the disclaimer about existing entities no longer being available and ceasing to work…I’m excited for all the improvements, but…
Is there any way to have future updates not render the majority of the existing entities to become unavailable and require exhaustive deleting?
It would ideal for all the existing entities to be kept and be forward compatible so we can maintain all the valuable history logs for all entities?
Should be fine now and in the future as my old code was all over the place and I spent a lot of time refactoring it and making sure it follows home assistant development best practises and integration quality scale.
Just want to say thank you for everyone providing feedback here and on Github.
It helped a lot to make improvements.
I spent a lot of time refactoring the code and file structure for this integration.
When I test the integration on my i7 8700k 32Gb RAM UNRAID server which has 1 Parity, 3 Disks and a Cache. My HA logs doesn’t show any warnings or errors for my system so when I release a new version, its interesting the see the issues that get logged and how the integration behaves on other systems with more disks, different chipsets, fans, cpus, motherboards, etc.
Hopefully the new code structure makes it easier for someone else to look at and understand how it all works and help squash more bugs and make improvements. 
3 Likes
Thanks for the additional info. Yeah it didn’t help re-re-reinstalling, I’ll have a proper look when I get some more time, nothing’s showed up in the logs so no errors at least.
It’s been a day or two as well of attempts so it “shouldn’t be cache”, but that doen’t mean anything in the HA world.
Where’s my manors anyway I was so focused on those switches I forgot to say thank you for that recent update! It’s amazing, all the docker stats are wonderful.
Good to know you like the docker stats. It’s a Frankenstein feature which won’t stay due to stability issues and relying on docker socket proxy container to work.
My favourite type of features. It’s been working solid here and I didn’t really mind the additional container. Be a shame if it went but I understand.
Was there a reason for using the container over the inbuilt ssh command docker stats? I see the rest of your sensors are SSH based so just curious as I’d intended to script a similar feature over mqtt a while back.
Edit: I have found some errors, something reading the parity check state and something else with a mount point so far. Still not switches have appeared on my instance but I do have a browser console error for not subscribing to 44 entitles on that page so it might still be cache. Will investigate more and I’ll enable debug logging and do some github issues for them when I’m confident it’s nothing I’ve messed up.
Th docker socket proxy container has direct access to /var/run/docker.sock and allows you to pull more container stats than running docker commands over SSH.
This integration works quite well and gives more options than the one I worked on.
1 Like
@ruaandeysel After reinstalling for the new version, I’m seeing a thousands of these errors and warning in HA logs:
Also did I miss something with Docker entities, it seems like they’re all gone since the prior version and it’s misreading my Docker service as not running, which is invalid.
Discovered a bug with Unraid version 6.12.x. When the new integration code does a validation check to see if VM and Docker service is running it fails on 6.12 but on 7.0 it’s working fine. So looking now into tuning the code to check for few extra things.
That should sort out the missing VM and Docker control switches.
(I’m aware in the old code it worked fine but it didn’t have validation checks which caused errors in logs when users didn’t use vm or docker service)
Logged as a bug and can find more info here
https://github.com/domalab/ha-unraid/issues/36
The invalid parity history line is the code that’s trying to read previous unraid versions of parity check records in /boot/config/parity-check.log that had less columns and info than the new versions of parity check. Not idea how to fix that yet but its logged as a bug and can follow it here.
https://github.com/domalab/ha-unraid/issues/38
Please log an issue for the Error getting stats for interface eth0.4 issue.
1 Like
Thank you for this integration! It is amazing.
My only concern is about granting full SSH access instead of limited read-only access to the HA integration.
I think this is a security risk as Home Assistant itself, on a typical setup, is exposed to the public internet.
It would be nice if there is a way to restrict the access to read-only instead of full management-level SSH access credentials.
Unfortunately, unraid does not currently support role based access control on accounts and I have not seen or found a way to use leased privilege access.
It would have been great if Unraid had an API we could use.
2 Likes
I would not expose home assistant to the internet, not even with letsencrypt and nginx add-ons.
If you need remote access to it, use a VPN. For example, all my mobile devices have full VPN tunnel back to my home network using Wireguard.
1 Like
@ruaandeysel thank you your attention on this!
I created a new issue for the errors related to the sub-interfaces:
Would love to see it monitor the docker vdisk size. Thanks for your work on this!
Edit: I’m dumb, its there.
1 Like
Awesome! I’ve been looking for a way to get my UPS power consumption into HA energy dashboard
Woke up to a nice surprise this morning, saw a notification under HA settings that there was an update for the Unraid integration for version 2025.01.08.
I just upgraded 
I would like to donate to this project to show my gratitude. 
1 Like