thewave

Thanks for the reply - you'll note that I mentioned I upgraded from one Quadro to a different Quadro that is two generations newer; which is partly why I didn't expect a dummy plug to be necessary

I recently upgraded from a Quadro P4000 to a Quadro RTX A4000 and I'm noticing some problematic behavior.. It looks like dockers that rely on the GPU do not function correctly. Looking at syslog, I see this error a few times per minute: kernel: nvidia-modeset: WARNING: GPU:0: Unable to read EDID for display device DP-0 Notably, if I plug the server into a monitor and reboot everything appears to function correctly. I don't have a dummy displayport plug on hand, but I suspect that may improve the situation. Regardless, this wasn't an issue with the P4000 and the server was perfectly happy functioning headless. Does anyone have any ideas about how I could fix this?

You nailed it. Adding cdn.jsdelivr.net to pihole whitelist fixed the problem.

Did anyone's dashboard icons recently break?

Is this fix meant to address that the WebUI is no longer accessible after installation of the plugin? Or is this fixing a different issue?

Also having the above issues; installing this plugin immediately breaks webUI access to Unraid.

I recently noticed that I had an issue where the mover was not moving any files. In the 'Main' Tab, clicking 'Move' would not cause any action; however going into the individual 'Shares' tabs and moving the shares individually would work as intended. I implemented the solution above, removing the CA Mover Tuning plugin, and the issue was resolved.

You're teaching me about both Matrix messaging as well as Pushbits right now; I just want to say that I appreciate your time. The changes I've just made: I switched the initial admin username back to 'admin' and I made the password simple so I didn't need to copy/paste it. The Matrix ID of the initial admin is set to the Element account (beginning with @) that I'm using to sign into the Element app on my phone. I've just left homeserver as the default of https://matrix.org I created a second matrix account on element with the name username 'pushbits' and a secure password. I deleted the old config.yml and pushbits.db files and copied the new config to the folder, then ran the container. I didn't change the permissions of the new db file. And... Woo! I went to Unraid Notification Settings and set the Full server base url to my.unraid.ip:8050 (8050 being the host port specified in the docker template), and then I pasted the token from the above screenshot, and it works! I got a notification. But at this point, what I understand is that I'm using the public matrix.org server for notifications. You're saying that you recommend I self host a matrix server to use for this?

Do I need to have my own matrix server, separate from the Pushbits container?

Let's pretend my domain name is abcdef.com I have Nginx Reverse Proxy set up with a proxy host of pushbits.abcdef.com set to forward to my.unraid.ip:8050. Is the initial admin just a local account and the credentials don't matter that much? Should I leave it as the default admin/admin? Or should these credentials match what I use to log in to Element? I have set the homeserver in the config.yaml file set to http://pushbits.abcdef.com and the container will not start:

Deleting the .db file and allowing pushbits to recreate it did help and allowed me to get further. I'm now getting an HTTP 500 Internal Server Error: And the logs: Notably, when running these commands the container is able to create the rooms in Element: Do I need to use my own IP as the homeserver in the config.yml? Do I need to use two separate Matrix accounts for messaging?

Using the admin credentials I get: Request failed with HTTP 500 Internal Server Error. The logs for this:

Hey, I tried the exact address as above and I still get the 403 forbidden error. This may be related to unfamiliarity with Matrix messaging. I tried setting up an account on element and I've copy/pasted the password to the config.yml file as well as the console, but I get the same error. At one point, element said I had an 'unverified session' but said that Pushbits didn't support encryption and so I could not verify the session. Do you have a different app that you recommend to use?

Hey, this is setup through Nginx Proxy Manager This is what my console looks like in the docker: I tried with both the admin credentials in the config.yml as well as the matrix credentials

I will PM you the full command because it includes my username and domain. Also, the readme says that the listening port needs to be changed in the config.yml. The container port is grayed out and can't be changed in the unraid UI; does the config.yml need to match the container port or the host port?

Seems I need to use the pbcli tool - is this done in the console of the docker? When I try to enter the console I assume it errors, because the window immediately closes itself EDIT: I changed the docker console from bash to shell, and now I can access pbcli; it says the supplied URL is invalid. Is this not the listening port? Do I need a reverse proxy enabled for this to work?

Have there been any other reports of discord notifications failing? I'm able to send test notifications just fine, but other notifications aren't working. The Local TLD was set before I enabled the discord agent.

So I got the container to start. It looks like the container couldn't see the config.yml because of a permissions issue; the /mnt/user/appdata/pushbits folder had an owner of 'UNKNOWN'. I manually changed that to 'nobody' and the container was able to see the config file and start. For unraid notifications, I assume the URL is unraid.ip:listeningport. Do know how I can get the app token?

Well, when I looked at the logs and I saw this error: 2023/07/28 22:08:18 /build/internal/database/database.go:52 [error] failed to initialize database, got error unable to open database file: no such file or directory I figured I must have to create the db file myself - so I tried to do that in the unraid terminal. Based on what you just said, I deleted that db file. Now, trying to start the container it says it failed to find configuration config.yml , even though I can see the file in the /mnt/user/appdata/pushbits folder. Maybe it'd be easier to remove the container and start over from scratch?

Yep, I put it in the main directory but the container failed to start. I wonder if that means the container didn't have permissions to create the pushbits.db file?

It looks like the default config.yaml is using sqlite, but when I tried to start the docker without doing anything it said the database file didn't exist

I'm trying to setup a Pushbits docker container to enable Unraid notifications but I'm struggling to get the Pushbits container off the ground. I'm struggling with the database as I have zero experience with SQL databases. The error in the logs right now is : 2023/07/28 23:45:24 /build/internal/database/database.go:77 attempt to write a readonly database [0.550ms] [rows:0] CREATE TABLE `users` (`id` integer,`name` text UNIQUE,`password_hash` blob,`is_admin` numeric,`matrix_id` text,PRIMARY KEY (`id`)) Any ideas? Does anyone know where a setup guide may be located?

I followed the instructions in the first post of this thread and was able to get Stardew Valley working without any additional tinkering or playing with settings. At first the steam link did not recognize this as a computer but after restarting the container the problem seemed to fix itself. This is an awesome container, and thanks to the creator for the work they've put in.

I use 2FA with my google account - how do I enable it for Tailscale?

So I’ve recently discovered and installed Tailscale and it seems great, but I do have some concerns about security. It seems that if someone has access to my google account they now have ssh access to my entire unraid server.. is this not a big concern?