emptyfish

Hello all - FCP is reporting MCE error(s) about 1x per week - I pulled the below from the log which I believe is the relevant event (there weren't any others). Does this require any action on my part? Thanks in advance! Jul 14 20:03:20 monitor_nchan: Stop running nchan processes Jul 14 20:53:35 kernel: mce: [Hardware Error]: Machine check events logged Jul 14 20:53:35 kernel: [Hardware Error]: Corrected error, no action required. Jul 14 20:53:35 kernel: [Hardware Error]: CPU:1 (19:21:2) MC23_STATUS[-|CE|-|-|-|-|-|-|Scrub]: 0x81480775c8fe4105 Jul 14 20:53:35 kernel: [Hardware Error]: IPID: 0x0000000000000000 Jul 14 20:53:35 kernel: [Hardware Error]: Bank 23 is reserved. Jul 14 20:53:35 kernel: [Hardware Error]: cache level: L1, tx: DATA

Memtest immediately went red alert 😀 thank you!

Hi all, I have been searching through the forums and have found some information, but nothing so far seems to fully resolve my issues. I keep running into problems, generally every 5-10 days where either the docker service crashes and is unrecoverable or the unraid webgui stops responding altogether. This server has been operating for several years without problems so I am at a loss on what has changed over last month or so. The hardware all seems ok - I did have a drive failure a couple of months ago but I had no issues replacing the drive at the time. When the docker service fails I have to delete the docker image file in order to get everything back up and running. If we webgui becomes unresponsive I have to do a reboot via SSH to get the system functioning again. That only seems to buy about a week or so before I get a repeat problem, often it happens when updating one or more docker containers via the WebUI. I am attaching diagnostics from today, if anyone has suggestions please let me know. Thank you! ansible-diagnostics-20241216-1221.zip

I am pretty sure the instructions posted on the Lemmy site would work if you have Docker Compose set up in your server. I haven't tried it but a stack should do the trick with a few tweaks. Someone has also already posted a request for Lemmy the LinuxServer IO team on it as well - they came through pretty quickly on Mastodon when the Twitter issues started; I am thinking they may be the group to get this up and running which will enable it to flow through to CA. Lemmy is definitely further along in development than kbin from what I've read on both. https://join-lemmy.org/docs/administration/install_docker.html https://discourse.linuxserver.io/t/kbin-and-lemmy-fediverse-reddit-alternatives/7667

Mine was either uBlock Origin or Decentraleyes if I remember correctly - I don't think I had any other extensions that would have caused the issue.

Did you check your Postgres Db and permissions on it? I had a problem with that early on. Manually creating the db solved my issue. If you are able to see the tables have been created then likely not your problem. Also have you tried creating/verifying your account from the container (terminal)? You might at least get an error message indicating the problem. https://docs.joinmastodon.org/admin/tootctl/

The instance you choose (or host) has nothing to do with your ability to follow people or content from other servers. The point of Activity Pub/Federation is to have multiple servers that can talk to each other, or choose to be an isolated community if it serves the need. Over the last several weeks every time the big social network does something crazy there's been waves of users setting up new Mastodon accounts. The bigger Mastodon instances are strained under the load - they are all community-driven so run/paid for by volunteers. Under heavier load existing users will also have outages/lag/etc. Small or individual instances are a good way to offset those user spikes - it's how Mastodon and similar networks were designed to work. I think if someone just wants to use mastodon then your point is well-taken, there are many instances out there with established communities that will allow you to get connected. It's also good to point out you can move servers - if you end up not liking your first choice you can always migrate your account without much effort.

I created an unraid template for the bytemark docker image that will send emails - it's available in the CA store: https://unraid.net/community/apps?q=bytemark#r Once you install just set your mail server in your mastodon config to the smtp container name and match the port to the container port (not host port). You can leave login/password blank in the mastodon config, just the host, port and from address. I've been getting emails without any problems. Fair warning to anyone using the wonderfall container... the latest build is already updated to Mastodon 4.0 which is pre-release. I wish I had waited for linuxserver.io to do their magic. 😀

This is an all-in-one thread for my template repository located here. Currently focusing on a few select containers that I have not seen published in Community Applications. If you're interested in helping I am a total novice on this process and would welcome any feedback. bytemark-smtp This image allows linked containers to send outgoing email. You can configure it to send email directly to recipients, or to act as a smart host that relays mail to an intermediate server (eg, GMail, SendGrid). Github: https://github.com/BytemarkHosting/docker-smtp Docker Hub: https://hub.docker.com/r/bytemark/smtp/ unifi-switchboard Full disclosure - this app was co-created using Claude Code... I know there are mixed opinions on this. I built this tool to solve a specific problem for myself and thought others might benefit from it. This app is built in Python and Flask; you can use it to toggle existing user-created UniFi zone-based firewall policies (user policies) from a web interface. If you are a UniFi user and already have your gateway (ex. Cloud Gateway Max) accessible via the internet you do not necessarily need this application. This functionality is also accessible from the Unifi mobile app inside your network. The primary benefit of this app is to prevent you from having to navigate through several menus in order to turn a firewall rule on or off. This app uses UniFi's API endpoints, so your Gateway must be accessible via LAN in order to work. The CA template requires an API key, you can generate on through your Gateway from Integrations > API Keys. You also need to generate a random secret key for Flask sessions, ex. python -c "import secrets; print(secrets.token_hex(32))"By default this app is only accessible internally - if you want to access from outside your network you need to handle through your existing proxy and also set TRUST_PROXY to true. I've tested in Cloudflared without any issues but proceed with caution of course.

I just got the wonderfall repository version running in Unraid after a couple of days' trial and error. I am planning to create a CA template for it. It's definitely not an out-of-the-box experience as far as app installs go - there are dependencies involved so it's never going to be for the faint of heart. I've never done CA templates before so stay tuned while I figure that out. 😀 Feel free to ask questions in the meantime if you're trying to work through it.

The real cause was me spending too much time tweaking late into the night and not enough time thinking. Firefox + addons were the root cause. I just needed sleep. 😀 Thank you!

Hi - I am trying to do a clean installation, new DB and new container for the front-end. Everything is set up correctly, but I am running in to an issue where I cannot log in via the web at all - I get the login page, I enter the user/password and it just keeps looping back to login. I know the ID/password are fine - I can sign in through the iOS app without any issue and I can see all the default files have been created. I don't see any error messages on the web and I have no clue where to even start to troubleshoot this - would appreciate any advice on where to begin. Thanks!

I have a (likely) very basic question on SWAG. I am running several internal applications inside my internal network. Using nextcloud as an example which I do have set up with a domain name, the docker port mappings show up like this: 172.17.0.1:443 > [unraid ip]:[port] This works fine until I reboot and then Docker may or may not change the IP of the container from 172.17.0.1 to something else. This breaks swag until I go in to the terminal and manually update the configuration file for nextcloud to the new IP address. It's not a huge deal but I feel like I am missing some very obvious step to prevent this from happening. I have tried mapping SWAG's proxy config using the container host name but it never works, it only seems to be able to resolve to the internal IP. Would appreciate any ideas on what I need to update if it's possible to fix this. Thanks!

I talked to the help Discord briefly. They seemed to think the Docker image is pretty out of date & initially recommended going Unraid > VM > Shinobi instead of Docker. I was able to get it working, though... I followed the update instructions on the Shinobi site (running the commands from the Docker console): https://hub.shinobi.video/articles/view/LTVqL3I8f8kIzsX I couldn't restart from the console, but restarting from the Unraid UI after running the update commands worked - everything came right back up. More importantly it seems to have solved the original issue - I am able to save presets/etc. Thanks again for the help!

Thank you for the response! I dug in to the DB and in my case it's not saving at all when I click "Add" on the preset Window. Furthermore, I tried manually inserting a Preset record into the Presets table and to your point it does not show in the UI at all. I traced the Add button and the POST sends data like this - seems like it's not providing any information: data={"monitors":[]} This is the Url that gets called whenever you click the add button: http://ip:port/7782fd1f9527265498580dada163bdb9/monitorStates/peRkP9V6hr/FrontYardMonitorOn/insert Even though there's no data getting inserted, the POST URL above always returns this: { "ok": true, "msg": "Inserted State Configuration" } It seems like the Preset/Monitor State is completely broken.