I think technical things are in the scope… Like Python makes some calls to the underlying system to mount network storage. If HA doesn’t know YOUR system it’s hard to make the correct system call and check you get everything correctly setup !
I have no clue how to use Network Storage to connect to Synology NAS. Whatever I try, I get errors.
There seems no connection to be made to the Synology as there is nothing in its log.
I tried both with my administrator account of my NAS and with a new account without 2 FA.
Also tried CIFS and NFS.
I have no problem mounting the folder on the Nas from my Macbook, so it must be something in HA.
With NFS I get this error.
Could not unmount NASRob due to: Transaction for mnt-data-supervisor-mounts-NASRob.mount/stop is destructive (mnt-data-supervisor-mounts-NASRob.mount has ‘start’ job queued, but ‘stop’ is included in transaction).
With CIFS I get this error
some but not all values in the same group of inclusion ‘basic_auth’ @ data[<basic_auth>]. Got None
Anyone a clue or description what to enter in the different fields ?
4 Likes
On any install method OTHER THAN HAOS it is pretty simple to mount smb or nfs directories in your underlying OS. So this wouldn’t make sense anywhere other than HAOS.
5 Likes
Indeed, makes no sense whatsoever
2 Likes
Supervised is supported, so a simple mount on a supported OS shouldnt be very much of a challenge 
1 Like
Because it is very simple to add on all the other installation types, especially in a docker install. In all other installations you as the user own the OS and it is up to you to define this.
3 Likes
You can already plan backup with automations 
But without retention strategy.
1 Like
Yes, I am. The HomeKit Controller is no longer working. It keeps saying ‘Initializing’ but fails to setup.
4 Likes
I stupidly did not save my docker-compose file with latest version before pulling. Thanks, and great release, zero issues on upgrade.
docker logs home-assistant
s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service legacy-services: starting
services-up: info: copying legacy longrun home-assistant (no readiness notification)
s6-rc: info: service legacy-services successfully started
Planning is just automation, and hopefully we get automations in Home Assistant. But retention is also something that is important. I don’t wan’t to manually delete old backups when I just realize my NAS is full !
Then, you can do your own mount if you have full control of the OS. No ? 
1 Like
A lot of improvements I see! Devices page looks great.
Only one minor dissapointment:
The new color pickers do not show up for LimitlessLED lamps. All of my lamps are just that… 
Thats not the point. Why not just make a nfs mount possibility in HA, thats nothing exotic. I’m not a developer, otherwise I would have made an Add-on myself
3 Likes
Let me explain:
The home assistant without HASS OS is just a docker container. Everything in there is isolated and run as an unprivileged user. To mount NFS, which is a kernel feature, the container either had to be run as privileged or as crzynik already said added to the docker run command.
Home assistant can do neither of these without having accesss to your host system while running standalone in docker. However, with HASS OS (or a Supervised install) there is an agent running on the host.
Home assistant can talk to it and do all the NFS things automatically for you.
TL;DR: No, they don’t wanna force you to use their OS; it’s a technical limitation and you can also install the supervisor without HASS OS.
Lemme know if something is unclear.
11 Likes
youre right man… i was enthusiastic about a new coool sounding feature i could already do since i run superawesomevised 
#facepalm
2 Likes
The way I have notifications set up is that If I’m home they play a TTS notification, however I don’t want this going off when I’m in a work meeting or on the phone etc. As such when either of those events are true I instead have it create a persistent notification. When I’m no longer in that busy state a automation then goes off using the persistent notification IDs ie persistent notification 1 2 3 4 etc to read out what the notification was. I also use this so it reads out the notifications that occurred over night as part of my morning routine.
The removal of persistent notification IDs by default in this latest release would completely wreck this set up and it’s not something I can really go without. Does anyone have a work around? Otherwise I’m going to be stuck on the former release. Wait for trigger would become problematic when multiple notifications go off. Is there a way to template a custom ID so it adds a 1 2 3 4 etc to the end of the custom ID if one already exists?
1 Like
/volumeX/path/to/dir
(X is obviously the volume number)
