Kees Fluitman

Hi Guys, I've been getting some regular crashes lately. Mostly when my server seems a bit loaded with applications. The specific error is Kernel Panic. But the thing is, since logs are going to the usb-drive, I have no way to see what happened or what caused it. Any tips on how to better diagnose it? I've attempted to go through the logs but to no avail. If you have any tips whether or not I should log to some external source, or if the diagnostics would help here? Ill post them. I added a diagnostics from a couple of weeks ago, when it also happened. Maybe it will help to give you a glance at my system. - memtest passed. - Ive set syslog to a separate share. So let's see once a crash happens. server-diagnostics-20240115-1739.zip

yes perfect. i did it that way as well. root@server:/boot/config# cat /boot/config/go #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & cat /boot/config/kees_aliases >> /etc/profile

hey. I've started noticing it didnt work anywhere. Any modal was throwing errors. The web dev console showed 507 responses I believe. That the server didnt have sufficient storage: I dont know why WebDAV though? Or if they mean memory...anyway... So i figured let's dive deeper. And multiple thing started to act weird. I then looked at my memory allocations, and usage. And it seemed there was no more free memory. At least, not real "free" memory. There was allocated memory for cache/buffer. but even that was starting to lower. At it's peak, memory used was 32GB, the rest cache/buffer (total 47GB). So i shutdown the system. and restarted. And now it's back to normal (25GB used, 11GB cache, and 8/10GBfree). I do already see a slow incrementation in my Grafana Chart. But i guess that could also be grafana itself at the moment? This eating of my memory started when i started to setup all my monitoring (Grafana, Prometheus, Telegraf, Loki, etc.) to play around. So I'll keep a close eye on that.

hey guys. I suddenly have an issue, where I click anything, compose up, or update stack, it opens the Modal window, and instantly disconnects. So nothing happens. Any ideas? I cant update or run most of my stacks now. I'll do it manually for now, but it's weird.

so to be more exact: - I create a script somewhere on my flashdrive (for instance /boot/config/aliases.sh) - I set this script to launch in the /boot/config/go file - all done? Or your way? You copy the profile?

did you find out if it's possible? Im a bit tired of unraid having these limitations :P.

hey. Old topic. But I was just seeing the same thing. My adguard-home is catching tons of requests that are repeated over time. All local requests, so they get a NX Domain response anyway! Apart from me going to have Unraid use a standard DNS address, i was still wondering...why does it make those requests anyway? They are all dns addresses of known open source projects, like pihole, traktarr, etc. Most of which i dont even have installed! So why make them? It literally looks like about 50 requests at once. To all known open source projects from A-Z.

bei post commands "--user telegraf:$(stat -c '%g' /var/run/docker.sock)" hinzufügen.

ive finally got it working properly. But not that properly... Im using nginx for reverse proxying it, i got it working. The only issue is, Im getting low bitrates, and only 360p max. It's also not getting better over time. I've opened ports 10000-20000 in my router and have tested calls with 2 or 3 instances on my home LAN. It just won't show in a nice quality. if anyone here has an idea? The bridge is connected fine, everything is. it's just giving terrible resolution, and only 15 fps.

I ran sudo tailscale up --advertise-exit-node But nothing happens, tailscale status doesn't show my server as exit node. Any tips? Also attempted this: tailscale up --advertise-exit-node --accept-routes --advertise-routes=192.168.1.0/24 But i still can't select the exit node from Tailscale Admin GUI

Im hoping to see a solution as well. I'd like to monitor connections going to my unraid mostly. VMs i can check, but my unraid can not be monitored at the moment.

Im in dire need of help concerning this topic. I've got Guacamole and would love to connect to my unraid host ssh. But it won't. I remember playing around with networking in the past, so might have ruined sth back then? Ive got two NICs, only 1 connected (eth0) Bridging is on for VMs (br0) Docker custom network type = ipvlan (selecting macvlan won't work, it will change back) creating a custom macvlan with parent br0 won't work either (it will start of course, but guacamole still can't connect to unraid ssh) I also somehow lost my custom br0 network along the way, and deleting the local-kv, and restart rc.docker only resulting in deleting my custom network and not recreating br0...with as consequence, me having to recreate my custom nginx network, and changing sth in all docker containers so they would properly restart again. Maybe they are related? Should I start a new topic? With proper diagnostics? I Would love to hear from you.

I can't connect to the unraid IP either. Anyone any luck? Custom Network (for my nginx), can connect to all VMs in LAN, but not to unraid IP. Should I enable host access to docker networks?

Hey guys. Suddenly my sudo -u#99 php occ throws me back unable to resolve userId. Worked yesterday...any ideas? Im on the newest unraid (updated last night). And 25 nextcloud. Php 8.0.25

are you still working on figuring it out? Ive been wondering as well, since some apps just have badly updated containers in the CA.

Yes I agree. The thing is, I access it through my own domain: owncloud.domain.com and it logs in properly. But then redirects with this error. Have you gotten the Web app working properly by the way? The prettier version of the interface? I got it working but it's really too slow imho. The pdf viewer is also not working in the web app. Ill go to the official forums if you dont know either. The desktop is working now. I figured it was weird that it was calling my localhost (my desktop app) on https, so after making it http it worked directly. With the Web app. I noticed that it's slow because for every document or folder, it sends an API request to the server. Even each single icon loads in subsequently. In the normal web browser, it calls on "GET /index.php/apps/files/ajax/getstoragestats.php--blablabla" So either i've got something set wrong, or it's bad design.

I can't get my desktop app to work. When I connect, it opens the browser, i authenticate and then it hits an error: SSL_ERROR_RX_RECORD_TOO_LONG it's trying to connect to https://localhost:63426blablabla

maybe an old topic. but I used this, and it scans properly, the folders are just of a different ownership. But Icant get the occ working if I go to my owncloud console, to my /config/www/owncloud and then use php occ. Maybe you know a quick answer, I cant seem to find out what Im doing wrong here. maybe my user is not www-data? edit: it's abc. thanks.

thanks. Owncloud feels much faster and more responsive than nextcloud already. Let's see if I can get the hang of it and then Ill stick to it.

Im Sorry I dont know why my shares would be the problem. I want a separate share for owncloud, because my appdata share is set to my nvme cache drive (so the files are not secure in case of crashes or power down). I want this set to my SATA ssd cache pool (2x 1TB ssd). So now I have a share: /mnt/user/owncloud I set the paths to: /mnt/user/owncloud/appdata/ownCloud/data (/data) /mnt/user/owncloud/appdata/ownCloud/ (/config) Yet still it gives me the same error and doesnt create all the files. Shall I send you the complete log file in pm? Even if I use the standard settings, with data in appdata/ownCloud/data, it's not working.

I've noticed they never spin down with nextcloud already...But my appdata is also a user share? Like all other files, unraid only has this. /mnt/user/appdata is also a share...so I dont get your point there. My appdata share is also set to prefer the nvme cache, which is not what I want from my data...I want it to be on a Yes SATA SSD cache pool user share folder. Is this the reason why it's not working? That it always HAS to be the appdata/ownCloud/data folder bind? cause as far as I can tell. It simply can't reach the config files. Since they don't exist somehow. even after just doing the standard setup. It's still not working and giving me the same error. Why should I put it in br0 and give it a static IP exactly? I mean, that can't really have anything to do with this type of error, but im just wondering. Any more ideas?

thanks for the quick response. I've tried your container, Can't I use my custom nginx network so I can use swag for reverse proxy? Even if I try br0 and a fix IP, the container shuts down after initial setup. Can't reach it on IP+Port with https. I've changed my data folder mapping to /mnt/user/owncloud/data (this is a share for owncloud, named owncloud). Otherwise I haven't changed anything yet.

I tried starting with nextcloud because it came up all the time. But I was very dissappointed with the initial troubles. Slow load times, some weird bugginess with automatic upload folders. I already lost way too much time with it...And all the extra functionality it has...I dont really need. I just want a cloud server. But I was wondering, I have a separate redis and mariadb and would like to have everything in there. Can I set up this container so I can optimize it that way? Or is that really quite unnecessary?

The containers just seem to still have a different order than they do within the folders. Should I have first ordered them, and then put them in folders? Because at least one container fails to start after the daily updates, because it's missing it's database I presume. ah I see, i had the old version installed. Not compatible. One app is still being started to early even though it s below my database.

Hey, I havent copied any certificates. I did succesfully reverse proxy my instance though. But I can't login on my app with my reverse proxied URL. It says Error: Please check that invoice Ninja v5 is installed on the server.