Marshalleq

And we are all very grateful for this, thank you.

I get a traceback when running aws-cli anyone else get it to work?

So I had your container disabled and had gone back to steamcache. I double checked why by re-enabling yours today. Basically with yours enabled I get the dreaded steam cannot connect to steam servers (or words to that effect). I'll go over my config again at some stage, but it's fairly defeating to have a game cache if it doesn't allow you to play any games! Thought I'd mention in case anyone else has any input / insight.

Ah yeah, I saw your earlier email and was about to paste it for you So that's good to know - we can probably use this to point something anywhere on the network.

I have been pondering a similar conundrum, and got to wondering if I could set up a vlan that matches the custom network created for letsencrypt. I'm pretty sure that'd be fairly easy - then if so it wouldn't be hard to set the same vlan up on other devices and do what you say. Otherwise, I'm sure lets encrypt can redirect to other addresses by a modification in the proxy-conf file. I think the Plex example explains how to do it. I think they call it host mode or something.

Thanks for the succinct explanation - somehow I missed that previously. I take it it's a binary which has no published source code - i.e. completely impossible to fix, or is it just that nobody has compiled it yet - i.e. possible to fix at some point? Thanks.

Awesome thanks, I'll take a look.

I can't remember if it was the same error message, but I did find once that the database IP address changed in the docker and I had to set it back. The same can happen with the nextcloud docker too - but you'd likely not be getting an error message like that if it had. So possibly the former, but it's just a guess. Also, that might be the docker log you've provided, not the server log - you'd need to look into the docker container or possibly in the docker folder on unraid to find the server log I suspect. That's what I'd do anyway - just to check all the logs.

Thanks the Lancache looks exciting - an updated version of steamcache finally! Just set it up and switched off my steamcache to test it out. Noticed the access.log isn't populating in the Nginx directory with hits/misses, is this no longer functional or did I misconfigure something? Thanks.

So far as I remember onlyoffice is a 'dumb' connection - no passwords or anything. It basically just points at the web server file system. So assuming you're using the onlyoffice container, I'd just wipe it out and reinstall as a first step (onlyoffice, not nextcloud). Then check the onlyoffice connector and try again.

Plex just put out an update - the changeling seems to indicate they've solve the issue, at least for the Nvidia Shield. If I were you I'd try it out again now and if it doesn't work, go jump on the update link and ask to add to the general builds too. In particular, I'm referring to this: (DVR) Only require 100 MB of disk space per recording for the transcode session directory. (nvidia-shield-issues#236)

Hey, yeah it's a good idea, I did consider that actually - I don't run nat reflection (I'm on OPNsense) - when I enabled it, it broke other more important stuff, so I run a dual homed DNS (It's manually set to the internal IP's for internal clients only). But to do what you're saying I'd need a triple homed DNS and the DNS server would also have to serve the subnet inside the docker containers separately - at that point I think it's not really a viable solution lol, so I gave up on that idea. I'm happy to have Onlyoffice in it's own docker anyway for now.

I found the problem, the instruction in the docker contradicts itself and I read it the wrong way. The below where it say's you should set the IP address to the hosts IP is incorrect. It's the dockers IP. If you use it in host mode, it would be correct, but the template defaults to bridge mode. Writing this here in case anyone else ever get's confused by it. Of course if I'd kept reading the line, I would have seen the conflicting information and set it correctly, but no I looked in a forum which gave the clue and then I read it! Lol. Hope it helps someone.

Thinking this through more, I wonder if the fix may be to set up a letsencrpt proxy setup, with it's own domain, but pointing at the same instance of nextcloud. Either way, I've gone back to using a separate only office docker instance for now with it's own domain, but with nextcloud 18.0.1 separate which works well. This experience has been good as I understand how it fits together a bit more now.

Regarding only office integration. I reckon the remainder of the issues are DNS / port related. As far as I know, onlyoffice and nextcloud need to communicate with each other across a port and ip / DNS name. What I mean is the 172 internal to docker addresses much communicate with each other - not the external DNS name, which would work if you were running host networking. Of course as far as I know, there is no way to tell nextcloud that nextcloud.yourdomain.com = internal to docker IP address, only the host IP address and the internet IP address. I suppose you could put a host file in there or something though. Anyway, so that would mean we'd need to put http://172.18.0.x in the advanced server settings (which I've done without success, I suspect because I don't yet know the port that it uses internally in this docker container). I assume I'm going to need to do http://172.18.0.x:9000 or something like that. That would explain why I keep getting can't contact host when opening onlyoffice in nextcloud. Anyone else agree?

Hi, has anyone got an active working steamcache? I set this up again and it starts, the DNS resolves, it even responds with the host name for the dns query listed in the instructions and all clients pointing at it resolve internet DNS correctly. However, nothing turns up in the logs either access.log or error.log. Seems a bit like the container is broken. Fresh install. Thanks.

Yeh, I had similar random errors like that too. I assumed it was space running out but can't be sure. Quite hard to identify as the stream has usually stopped and the files are cleaned up I guess.

I'm pretty sure the nextcloud pipe is very slow - it's certainly not my internet speed - which is 1Gb Up and Down (real world is 900/500). So i'm waiting for a proper fix - manual does it for now - though this only covers the downloading issue, it still doesn't work. On that note, has anyone on here gotten Nextcloud 18 working with the 'built in' Only office in this docker?

You mean the 'built in' onlyoffice, that's not actually built in? I've just tried upgrading from 17-18 to get it all in one docker. It hasn't gone so well. So far I've figured out the following: 1 - If you have the onlyoffice connector app installed (confusingly just called onlyoffice, but it isn't onlyoffice) remove it and re-add it to clear the settings - make sure they're cleared (or possibly just make them blank and save it - not sure). 2a - There is a bug that means you can't install onlyoffice (the community document server) via the nextcloud app store - to fix that, I manually downloaded it from https://github.com/nextcloud/documentserver_community/releases/download/v0.1.5/documentserver_community.tar.gz and extracted it into the apps container (you have to go into the container - docker exec -it nextcloud bash and issue wget then the url - place it under the www nextcloud apps directory and extract it there - it then shows up as an app. 2b - Apparently if you installed this after a fresh connector install, it will autodetect the settings - mind kind of does and misses off a few characters - no idea why. 4 - Enable the community document server - supposedly it's supposed to work as easy as that, but for me it doesn't - it doesn't like the url. So I'm stuck here because I'm not yet sure how it's supposed to work (I used to have it in a seperate docker behind nginx / lets encrypt proxy, which may be part of the problem). Annoyingly the separate docker approach also broke with the upgrade. I've read there are some permissions issues, but there error I currently get is bad health check status (in the onlyoffice connector), so yet to work that out. It might be something to do with the nginx proxy, but until I figure out how the integrated onlyoffice is meant to work I can't do that. I have to say for a feature that is supposed to be so easy, it sure is bad and has really bad docs too. After this, I believe I will have permissions issues to fix and possibly some others. So that's the steps so far, hopefully you can add to this! Marshalleq

This is why I disabled RAM transcoding and went back to doing it on an SSD. I have 128GB RAM, so it's not like I don't have options. I think what was happening is that the RAW stream of the show was very big and for some reason was trying to play a full copy in the ramdisk. I think I did validate at some point that's what it did (there's some commentary around it earlier in this thread I think). There needs to be a way to separate the two out and then it would work I suppose, but I didn't look into it much further - too many projects on at the moment.

Does anyone here know if the ZFS samba sharing uses the Unraid Mac OS Enhanced Capability as set in the GUI? I assume it does, but I'm particularly interested in xattributes, but there's actually quite a lot that can be configured via vfs_fruit and it's important to get right.

@steini84 I've been having a problem where snapshots don't delete (i.e. a plan that says keep snapshots for 10 minutes for 2 hours, then a lower frequency ends up creating endless 10 minute snapshots) and have logged a report on GitHub here. That got me to wondering if I was running the latest version (which probably they will ask). Your compiled version says it's 0.dev (znapzend --version) whereas GitHub says they're up to 0.19.1. I think yours is actually the same since the 19.1 came out in June last year. Can you confirm? Thanks.

I’m just going to let this go now. Sent from my iPhone using Tapatalk

Without clicking the link to check it’s been widely commented on one of those that he doesn’t know what he’s talking about when it comes to stuff outside of the kernel. As in there were many incorrect statements in it. Like I said, if you look for the negatives. Sent from my iPhone using Tapatalk

I’ve posted about it here twice already. There are already other spots on the internet. Look up Ubuntu for example. I’m sorry that I’m not being overly helpful but I don’t wish to start a whole discussion on something and end up in a whole opinion piece that neither of us can win in the wrong thread. . But just google it. I found heaps. Sent from my iPhone using Tapatalk