erf89

Ahh this is great, just tested and all working, thanks! Great project this, I was testing it out with Tinyauth the other day to SSO apps that don't have OIDC

Just tested that - looks like it's appending the same ports to my domain as I use for non-reverse proxy access. So I have :84 and :447 set as HTTP and HTTPS WebGUI in Unraid, but when I add unraid.domain.com, it's adding it as https://unraid.domain.com:447/graphql... where my reverse proxy doesn't need the port. It also doesn't seem to like IP's, I tried adding 192.168.50.5 and it didn't add anything to the IDP. (I tried it with a space after the first entry and also just on it's own in the Allowed Hosts field) - maybe if it was a field that doesn't append ports, tooltip could be to add the host:port and then the plugin is only appending /graphql/api..., for me that would be "unraid.domain.com 192.168.50.5:84 192.168.50.5:447" I don't know enough about the Unraid API, but could it reference the entries from "Allowed OIDC Redirect Origins"

Yeah I've got the Grant in my ACL, would be good to have that persistence, although not super important!

Came here to ask about persistence as well - I've added a couple extra redirect URI's in the TSIDP UI (LAN IP and my FQDN), they didn't persist when updating to the latest version of the plugin so I've gone back in and added them - will they now persist on future updates/reboot of machine (barring any breaking changes by Tailscale!)

Thank you - not sure which container it was, I had a look through before the reboot, and nothing had an uptime of "seconds", a few that were 4 hours uptime from the docker auto update, but at least that's resolved for now, I'll keep an eye on it.

Hi, Just had this warning in Fix Common Problems /var/log is getting full (currently 57 % used) On checking the syslog, I can see a lot of repeat warnings: shfs: fuse warning: munmap(0x149cc41ac000) failed I have attached my anonymized diagnostics, if anyone is able to help? An earlier post from a google search suggesting not using NFS shares and only SMB, but I don't have any NFS shares enabled.

Hi I upgraded to a Jonsbo N3 at Christmas, so looking to pass on my N2 to anyone who is interested In the UK usually retails for around £155 on eBay etc. I used the case for around 11 months £100 + delivery, or pickup in Cheshire East I will add some pictures soon, the case is currently sat in the storage unit!

Just woke up this morning to a Fix Common Problems alert that my server was out of memory. Looks fine this morning, sat at around 80% where I left it last night, but I've downloaded the diagnostics anyway! No idea where to even look in the diagnostics to see what caused the error. farrosphere-diagnostics-20241208-0712.zip

All working, thanks!

Not sure if this is a bug when using in 7.0.0-rc1, but I can no longer see the Tab Name or URL options. I had an existing tab setup so that is preserved, but my tab for User Scripts I just created is called Custom1 and I can't seem to rename it

Saved me from rolling back/pulling my hair out! Thanks! 😄

I have the CA Mover Tuning plugin installed and setup to run at 85% full - was working no issue until today, I've updated to Unraid 7 beta, and now I can't get it work, and I also can't get manual mover to work either (I set "Move Now button follows plug-in filters" to No), also tried using the Move Now button inside the Share settings page, also no luck! Snip of syslog below, happy to provide any other diagnostics detail! Jul 8 00:55:21 emhttpd: shcmd (170): /usr/local/sbin/mover start |& logger -t move & Jul 8 00:55:21 root: Starting Mover Jul 8 00:55:21 root: Forcing turbo write on Jul 8 00:55:21 kernel: mdcmd (43): set md_write_method 1 Jul 8 00:55:21 kernel: Jul 8 00:55:21 root: ionice -c 2 -n 0 nice -n 0 /usr/local/emhttp/plugins/ca.mover.tuning/age_mover start 0 0 0 '' '' '' '' '' '' '' '' 50 Jul 8 00:55:21 move: Log Level: 1 Jul 8 00:55:21 move: /usr/local/sbin/move does not exists, creating soft link Jul 8 00:55:21 move: ln: failed to create symbolic link '/usr/local/sbin/move': File exists Jul 8 00:55:21 move: mover: started Jul 8 00:55:27 move: /usr/local/emhttp/plugins/ca.mover.tuning/age_mover: line 206: /usr/local/sbin/move: No such file or directory Jul 8 00:55:27 move: cat: write error: Broken pipe Jul 8 00:55:27 move: /usr/local/emhttp/plugins/ca.mover.tuning/age_mover: line 119: /usr/local/sbin/move: No such file or directory ### [PREVIOUS LINE REPEATED 4 TIMES] ### Jul 8 00:55:27 move: mover: finished Jul 8 00:55:27 root: Restoring original turbo write mode Jul 8 00:55:27 kernel: mdcmd (44): set md_write_method auto I should add that as well as being on Unraid 7 as of today, a few days ago I also changed my cache pool to zfs from btrfs, however the Mover getting everything back onto the Cache at that point worked fine, and today is the first day I'm hitting my 85% again.

Just installed steam-headless but can't get it to work - WebUI gives me the noVNC screen but when I click Connect, I get "failed to connect to server". I have a Radeon 6400 and have installed Radeon-TOP, and can see the GPU on my Dashboard. (GPU was previously bound and passed through to a VM, I have unbound and rebooted unraid to get it back to normal and see the GPU on my Dashboard). The first time I installed steam-headless, I ran it with --device=/dev/dri in the extra parameters to use my Intel iGPU, however because I only want it to use the Radeon GPU I have removed that, no joy, and then completely uninstalled, removed appdata folder and reinstalled without ever adding /dev/dri, and still the same issue, can't get it working. In the logs I can see the below: [ /etc/cont-init.d/60-configure_gpu_driver.sh: executing... ] **** Found Intel device 'Intel Corporation Alder Lake-P Integrated Graphics Controller (rev 0c) Intel Corporation Alder Lake-P Integrated Graphics Controller (rev 0c) Intel Corporation Alder Lake-P Integrated Graphics Controller (rev 0c)' **** - Mesa has already been installed into this container **** Found AMD device '03:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 24 [Radeon RX 6400/6500 XT/6500M] (rev c7)' **** and then I get a load of WARN: exited entries in the logs Anyone able to help?

Fix for now is to rollback to the previous version: binhex/arch-delugevpn:2.1.1-6-03

Same here for me!

For anyone attempting to install Cryptpad needing to use different ports (3000 is used by Grafana by default which a lot of us will already have installed!), I struggled at first but wanted to share how I eventually got it working! Decide ahead of time which ports you want to use (I went 5000, 5001, 5003) as this will speed up the setup process. I followed the instructions in CA Apps to change permissions, make directories etc., then installed the container, but amended the pre-set Ports. You will need to REMOVE Port 1, Port 2, Port 3; and then re-add them, matching the Container Port to whatever Ports you intend to use (so Port 1 should show 5000 in each box) Install the Container, and then we can finish with the amends to config.js. Amended the config.js file as per instructions, ensuring you Uncomment the httpAddress setting and set as 0.0.0.0 I then uncommented and amended the three pre-set ports (3000, 3001, 3003) to be 5000, 5001, 5003. After all this I then was able to get into Cryptpad WebGUI successfully and create docs.

Ahh yes, that's working now! Thanks for your help!

Thanks for the quick response! Just did a force update but still not working, the Logs are very much the same though, says it's started the metrics server

Just installed the cloudflare tunnel and all setup with hostnames now! I'm trying to add the metrics to Prometheus, I've gone into the docker template, added a specific port, and also added the Port to the container - looks positive when I Apply, I can see the mapping on the Docker page, and the Logs tell me that the metrics server has started on that port, however when I try to add to Prometheus, or even just navigate to the path, it doesn't work Anyone know where I'm going wrong?

Inspired by a Linus Tech Tips video two years ago where they built a Home NAS in a Jonsbo N1, and subsequent months... years... of Googling, finding the Jonsbo N2, I decided it was time to finally time to dive in and build my first NAS. Prior to this build, I ran Plex on my main daily PC which was essentially on 24/7. That PC is now only on when I'm actually gaming or working. A lot of the parts were inspired by a previous UCD post, as well as a video of an all-AliExpress NAS on YouTube that I now can't seem to find. OS: Unraid 6.12.8 Basic (since upgraded to Plus and 6.12.9) CPU: Intel i5-12500H (CPU part of the motherboard) Motherboard: Erying G660i (CPU part of the motherboard) RAM: 2x 16GB Corsair Vengeance LPX 3000MHz DDR4 Case: Jonsbo N2 PSU: Dragon Storm 600W SFX SATA Adapter: NVMe 6-port SATA Riser Cables: 6-port SATA right angle cable Cooler: Noctua NH-L9i-17xx Fan: Noctua NF-A12x15 GPU: AMD Radeon RX6400 Low Profile Cache: 1x 1TB NVMe SSD Parity: 1x 8TB HGST HDD Storage: 3x 8TB HGST HDD, 1x 8TB Seagate BarraCuda During build I replaced the case stock fan at the back (exhaust for the HDD's) with the Noctua NF-A12x15, and also plugged this into a motherboard header instead of the included backplate, using the Fan Auto Control plugin to control the speed. Had the build running for around two weeks now, have since been playing around with various plugins and settings and found the SR-IOV plugin for the iGPU to be able to passthrough to VM without binding the whole thing and losing access to it in Plex for Transcoding. I added the RX6400 about a week after build just because it was cheap on eBay, not really sure what to do with it yet, I'm thinking maybe have that bound and passed through to a VM and have the server hooked up to the TV, maybe running Pop!_OS with Steam and a few others. One annoying thing with the motherboard is that all USB ports are on one controller, so can't bind to a VM. However, in my potential use case of hooking up to a TV, it's enough for me to leave a dongle for a wireless Dell keyboard/mouse plugged in, and probably do the same with a Bluetooth adapter and just have those two USB devices passed through, and I could then use a Bluetooth gamepad. The BIOS isn't the friendliest to navigate but eventually all setup with virtualisation, both GPU's enabled and XMP for the RAM, I even have resize bar enabled but not yet tested. Currently using this for Plex and the -arr apps, as well as running a DietPi VM with HAOS. Installed Tailscale to access remotely anything I don't want to open up to the internet, and running NPM to open up some of the -arrs using CloudFlare for outside access. PiHole Local DNS setup for all Containers internally. I still have 1x NVMe slot going spare, so may add another 1-2TB drive, either to boost the cache pool, or perhaps to leave unmounted and passthrough to Pop!_OS with the discrete GPU. Also scope to add 1x 2.5" SSD to the build if I wanted to, the case has mounts for it, although those drives are usually more expensive than the NVMe ones these days! Top of the case (missing the GPU): Front view, currently under my desk: Homarr Dashboard shows most of the Containers currently running: