Ryonez

I'm happy to see any steps towards docker-compose support. So many use it now.

The USB Flash Drive backup lets us keep more control over our backups. Deprecating this in favor of a method that only backups to the cloud is a mistake in my opinion. Would you do the same with appdata if unRaid added an option to backup appdata to their servers? This isn't even accounting for the fact that the backups made by unRaid aren't encrypted currently. I ask you to please reconsider this stance, maintaining the ability to create controlled backups of flash is important.

Could I ask why?

Maybe this would be better making a support thread for. This one is a feature request thread.

I see, thank you for the info.

Recently I've noticed that fields I remove from templates have started showing back up, and will continue to return whenever I try to remove them again. How do I prevent this behavior?

Now I've opened my eyes and actually read you post properly, I see that I missed a bit. Sorry about that. Gave that a go and it seems to work. Resetting the configuration is nerve racking though, everything is saying it'll wipe stuff.

Hmm, isn't the key Hmm... Does the trail key have to be online all the time, or just for registration? And could some sort of "emergency key" system be set up? Have it work like a trial, except this requires an old key to be present/given and will only last 7-15 days. Then the old key can be added to the "live" blacklist (prevents multiple emergency keys from being issued), while giving a licence holder the ability to get a server back up asap whille they what for a proper one to be issued. The keys should have the dates they work in them, so it would work for an offline system to. It would work, if I wanted to manually go through and set every thing back up the way it was. Drives, docker, plugins, vms... It's a bit much.

Interesting, wasn't aware the blacklist was baked into the system. I wonder if there's a way to still alleviate the issue I'm having without to much disruption to the system in place. Maybe allowing trail keys for existing unraid configurations. I believe it has to ask the unraid servers for a key to be issued, so it shouldn't be to hard to do a check there. Is there a reason for the current prevention there? I'll agree it's outside of a trail scenario, but it would allow us to bring the server back up, for a while. I wouldn't even care if it was for a smaller time frame as long as it'd hold until a replacement key is issued.

That's a brilliant idea, didn't even think of that! I tried, but it's throwing this: `It is not possible to use a Trial key with an existing Unraid OS installation.` So, can't do that.

I've just had a usb key die on me for the second time in the space of a year. The registration system allows for automatic key replacement once a year. If you need more, then you have to contact support. I personally am finding this frustrating right now. I've sent an email, and the form said it could take days. That's days my plex server is down. Days my automation system is down. Days my email server, discord bot that servers several guilds, database systems, wikipedia and notekeeping, git services, everything are all down. And not because I'm waiting on hardware, but because I need to wait for another key for software I've paid for, that has an automatic key replacement system. And sure, I'm not some big company, this doesn't affect others to badly (bar the bot that's down), but this is important to me and a few others. Why is there such a limit? The old GUID's are blacklisted, and I'm not seeing a benefit with the once a year key replacement.

Is there a way to connect pi-hole to two networks? Atm I have it set up with it's own ip address, however I want to secure the admin interface. I can't do that though because letsencrypt is in a secure docker network for the containers, and just pointing it to pi-hole will send traffic unencrypted over the lan network. I made an issue on the pi-hole docker github regarding this issue here.

Squid is meaning using pi-hole as the DNS server unRaid uses for itself. Though looking at the replies for it I don't get why this is pointed out. I don't use it for my server, and that's because I host it on that server. I might personally try it if it was hosted off the server, but I could see there being issues where it might block something unRaid needs by accident. It's unlikely you'll need to filter the traffic for unRaid though. Pi-hole is fine to use as the DNS server for wireguard, though there's some tweaks you need to do if you're hosting pi-hole on the same unRaid server as wireguard.

Is it only possible to have one active tunnel at a time? I've been trying to set up a second one to create a game network. I have another one, an admin network that has docker access that's working fine. But this second one I can get the client and server to handshake. Any idea what the issue might be?

Good to know, thank you for the clarification!

I'm a tad confused, why do we need to join their poll if we can vote here? Though I'd like to point out, if this is being considered a serious poll being taken on board by the unraid team, I'd much rather see it handled completely on your site, here on the polls. I've left facebook long ago. Not an issue for most, I agree, but for myself I don't wish to use a social service that has such massive disregard for it's members.

Was just doing some work on the server, and noticed a .csv file in /mnt/user. It's full of entries like this one below: 4,~DFEEA37C5D3D4ADFF5.TMP,/storage/Backups/Windows Reinstall/barro/AppData/Local/Temp,16,100 It seems to be structured, but I don't know of anything that would make a file like this, much less place it where it is. Does someone here have an idea what this could be, and if it's safe to remove?

Just set your "documentMaxLength" to 2000000 in the config and remove the CMD_documentMaxLength from the template. In fact I have almost everything done in just the config file. Going to let the others know, the docker image is still broken, though I guess the arm build is okay. Just to clarify, Version 1.4.0-ls11 is the last build I can get working.

Alright, thank you. I might poke around this subject more myself later.

I don't know. I'm a little reliant on others knowledge here as I'm by no means an expert and what I do know is rather patchworked.

Yup, I linked it with a edit shortly after. Sorry, I didn't think when I posted it to add it.

The lack of `Docker Isolation through Linux namespaces / subuids`. I've been given the example over here that if someone breaks out of a container when they are root, they are root on the host as well.

Any chance there are some updates? The lack of docker isolation is actually starting to worry me as I do have a few external services I host on unRaid.

Okay, did some testing and I believe the current docker image is broken. Version 1.5.0-ls12: Image broken. Seems to be missing the correct public/build files to run. Version 1.5.0-ls11: Image broken. Throws error in console: `s6-applyuidgid: fatal: unable to exec ./node_modules/sequelize-cli/bin/sequelize: No such file or directory` Version 1.4.0-ls11: Works. Has insecure pdf support, but is functional. My instance isn't public, so I'll fall back for now until it's been looked at and fixed.

I can't really test that. Beside using oauth, I'd have to change the config to edit the domain and such. When attempted as is, firefox throws security errors and won't load part of the page.