aldumil

Hello! I upgraded my motherboard recently, and since that upgrade, loading times through the reverse proxy have been very slow. Local load times are unchanged, just as fast as ever. Nothing changed except for the motherboard. I'm not really sure of how to go about troubleshooting the problem. Everything works as intended, but loading things through the reverse proxy (nginx prox manager, Cloudflare) takes forever. Would appreciate any input or guidance!

Hello! I recently upgraded my server to a new motherboard. Unraid itself runs fine, but I have a handful of services that are behind a reverse proxy for family/friends and these sites now load super slow when I didn't have this problem before the upgrade. I had some issues with connectivity and such with the new motherboard, but (as far as I know) those problems are now solved, but the slowness remains. I used https://pagespeed.web.dev/ to analyze one of the services and the main problem was: What can I do to try and troubleshoot/trace the problem? I'm using nginx proxy manager, but nothing in that changed so hopefully that isn't the problem. I should also say that the Unraid local webUI loads as snappy as always, no issues there.

I was, but I've removed it to see if it helped. It did not. Is there something I can/should do to make sure any issues it was causing get removed as well?

Just to see if it helped, I just took the PCI card out. I just want to get the darn thing running normally, then maybe I can circle back to try and get an additional NIC working. One of the motherboard ethernet ports might not be working? When one of them is plugged in I still get the "no cable connected" warning message, but the other port makes that message go away. However I still can't access the GUI over LAN, or internet obviously.

I feel like I'm taking crazy pills...I just keep changing the same few things and rebooting and nothing changes. It has somehow gotten worse though, now I can't even access the web GUI over LAN. It says it has an ipv4 address, but I can't connect to it from any computer. Fresh diagnostics attached....I'm really losing my mind here. syslog tower-diagnostics-20250323-0651.zip

I've found in my searching that unraid will show that message if one of the interfaces doesn't have a cable attached, even if the other one does. I guess that isn't really a clue to my problem.

I tried that on both interfaces and there isn't a difference. New diagnostics attached here just in case? Another thing that may be potentially relevant, on the network settings page is keeps displaying "Interface Ethernet Port 1 is down. Check cable!". This is clearly not the case because I'm looking at the page over a LAN connection. I did try replacing the cable just in case, but that didn't change anything either. Additional information: I found this thread and tried deleting the wireguard folder with no effect. I also have tried booting in safe mode, and that doesn't help either. tower-diagnostics-20250323-0823.zip

I did some upgrades/sidegrades on my server, booted it back and everything was working fine for a bit. Only issue was that I had to use the network interface on the motherboard because the NIC I installed wasn't working (it kept coming up UPv4 not set...). I've switched eth0 and eth1 back and forth, active/backup, made one inactive and rebooted with both plugged in....the issue now however is that whatever I have done in my 'troubleshooting' (changing settings, rebooting rinse repeat, hoping something good happens) is that now I apparently can't get the server on the internet at all. LAN access is fine on one of the interfaces (the motherboard one), the other still comes in as IPv4 not set, but I can't get the server to see the internet at all. I'm really at a a loss. I feel I've probably done something dumb in my fumbling, but I would really appreciate anyone who could take a look at my diagnostics and tell me if they see anything wrong, or how to fix whatever it is I've done. The NIC was working fine before I made the hardware changes, for what it's worth. The hardware change in question is a new motherboard. New to me anyway. tower-diagnostics-20250323-0123.zip

I was having the same issue, what worked for me was to remove the path, and then add a new one with the container path /app/config and the host path /mnt/user/appdata/glance/. Must have been something in the original path that was causing that issue, who knows.

On Unraid 7.0. Using Compose Manager to run an instance of Dawarich. I think there is a community application template is available now, but I've been running it since before that was up. I have had this issue in the past, but I thought it was fixed. The most recent update the developer switched from `postgres:14.2-alpine` to `postgis/postgis:14-3.5-alpine` and the issue is back and I'm not sure how to fix it now. I've attached a copy of the compose.yml to look at. I can chmod 777 (I know not a good idea, just trying to tracking the issue down...) the appdata folder and restart all the containers and it seems to run as it should, but after some time it invariably starts throwing things like this: I'm guessing the way I have the folders mounted is somehow messing things up, but I've been trying to read up on GUID/PUID/UMASK stuff that I see on some other templates to see if that can be incorporated here to give lasting permission to the containers, but I haven't found the magic bullet yet. I would appreciate any pointers! dawarich-update.yml

There isn't I don't think, this is the contents of logs folder on the flash drive: (the syslog.zip has a log for July 18, 2024) syslog syslog-20240416-092943.txt syslog-20240429-123922.txt syslog-20240510-130521.txt syslog-20240520-074515.txt syslog-20240619-191542.txt syslog-20240718-135950.txt syslog-20240718-162137.txt syslog-20240722-135053.txt syslog.zip tower-diagnostics-20240722-1350.zip tower-diagnostics-20240917-0747.zip

Edit: I think I meant ngx error, not nginx? I don't know if that is relevant. My Unraid server seems to have crashed overnight. The computer itself was still on and seemed to be running fine, fans all on, hard drives whirring, etc...but no services were accessible, and the monitor couldn't get a signal. I had this same problem a few months ago, I remember doing some research on the log output, didn't really get anywhere and then the problem went away on it's own. I had the setting to mirror syslog to flash on shutdown, but since I had to do a hard reboot I don't think it did that because their aren't any logs on the flash drive from recently. I've set it to mirror all syslogs to flash for now, which I think means if this happens again I'll have some logs to review? Is there anything else I can do to try and troubleshoot? Attached the diagnostics from right after getting it back online, just in case. tower-diagnostics-20240917-0747.zip

Yes, the list was empty. I had only set it TestMode to "Yes" for a test run, thank you though. TestMode doesn't seem to work though, it just ignores the setting and moves thing like normal.

Ah, so schedule the Mover is what trigger the Mover Tunning plugin to run? I thought the settings were independent of each other. What I had setup would run the Mover when cache got over 70% and it would run on Sunday at 1:00 regardless of capacity. So if I want the Mover to run when it hit 70% I should set the Mover schedule to be like...run every few hours?

Like this:

Sorry! I edited my post to add that while you were typing! I couldn't figure out the spoiler thing to do it fast enough.

Hello! I am having a problem with Mover Tuning plugin, I'm not sure if anyone can point me in the right direction because of how little information I have to share, but can't hurt right? Short story, it doesn't seem to work. I enabled logs and set the Test Mode to yes, but Mover seems to have just worked as usual as my cache is empty now, and the /tmp/Mover folder doesn't have anything new in it, just these files, but they all look to be empty, though I'm not very experienced using vi so maybe I'm doing it wrong. Or maybe there is a better way to access the logs in that folder? Cache_Tuning_2024-06-23T010001.list Custom_Mover_Tuning_2024-06-23T010001.list Custom_Cache_Tuning_2024-06-23T010001.list Mover_Tuning_2024-06-23T010001.list Would appreciate any help! As it is now the cache will just fill up completely if I don't manually trigger it. I'm including this in case it is helpful. Also, I had to add the filename because if I copy/paste from the FAQ post, it doesn't work. Maybe that is a problem also? I wanted to put this in a spoiler thing but could not figure that out. sh-5.2# cat "/boot/config/plugins/ca.mover.tuning/ca.mover.tuning.cfg" moverDisabled="no" moverNice="0" moverIO="-c 2 -n 0" threshold="70" age="no" sizef="no" sparsnessf="no" filelistf="no" filetypesf="no" parity="yes" enableTurbo="yes" logging="yes" force="no" ignoreHidden="no" beforeScript="" afterScript="" omovercfg="no" movenow="no" testmode="yes"

It's just a link to the Dockerhub page, unfortunately. Tried searching around but didn't have much luck.

Installed the Crowdsec - Cloudflare Bouncer container via the app store. The log gives me "the LAPI can't be reached" and I think it is because it IP is showing as 0.0.0.0:2112, while the Crowdsec container is 172.17.0.76, so I can't do 'localhost' or 0.0.0.0, but putting 172.17.0.76 doesn't work and putting 192.168.54.68:8080 (or :9767) doesn't work either. I'm not sure why the Cloudflare bouncer container isn't getting a 172.17.0.x IP assigned.

Just a screenshot of the pi-hole exporter not running, I can't find any logs that have any information about this. I checked the syslog, the prometheus container log, as far as I can tell their aren't plugin specific logs (?). I suspect the issue is on my end, if I got to the pi-hole address (192.168.54.101:80) I get a 'placeholder' page for lighttpd. My pi-hole is running in an LXC on proxmox, the API address is http://192.168.54.101/admin/api.php?. Is that all typical and I've just messed something else up somewhere? Thanks for any direction you can offer!

Hello @ich777! What a wonderfully helpful walkthrough/tutorial! I appreciate it! Do you still have a link to a 'buy me a coffee' thing? I'd love to show my appreciation, I thought I saw one at some point, but can't find it again. I have the adguard exporter working perfectly, but I can't get the pi-hole exporter to work in addition. Should I not expect them both to work?