Saldash

Hi All, I wrote some time ago that I was experiencing what appeared to be somewhat random system halts where the only way to resolve it was a full power off as the plug, wait and reboot. There would be no apparent cause for this but for a while it stopped happening, so I let the issue go. It has recently started flaring up again with a vengeance, so much so now I have a smart plug attached to the machine so I can remotely cut power and reboot. I still don't know for sure why this is happening and I'm not sure where to start. I've tried looking in the syslog but it only appear to start from the time the system came back online after it's last reboot, which is no good. I think there may be a correlation between the system halts and disk usage - A parity check is scheduled for 2AM every Sunday morning, and that is roughly when the machine stopped responding. Looking at the power monitoring data my smart plug provides, you can see the halt, shortly after 2AM when the parity check starts, then a long stretch of idling at ~80 watts before I cut power and rebooted earlier this evening. My machine typically idles about the 50 watt mark as I'm the only user, it serves very light NAS duties but does handle downloading and local plex media streaming. All my drives are reporting a healthy state and I've never seem the error count on them above zero. I've run self tests on all my drives but they all report completed without error. I'm not sure what to do next with this other than bite the bullet and move to a Synology or QNAP box 😟 EDIT: I Turned on the SysLog server and pointed Unraid back at itself to log events, then triggered a Parity check. I got about 0.8% before the server crashed. It does not appear to have logged the issue though as the crash occurred at 00:34 tabris-diagnostics-20230108-2347.zip

Hi all, I've started getting periodic systems halts that require a full hard reset to resolve. The system boots and appears to start normally until I get to entering my array password where after starting the array the UI appears to hang with the unraid logo pulse for a good minute at least before the UI becomes normally responsive again. I run weekly array health checks that have not reported any issues and after a fresh boot up, I cannot see anything in the system log that would indicate the source of the fault (not that I'd know what I was looking at to be honest). I've uploaded the diagnostic file in the hopes that someone might be able to point me in the right direction to getting this sorted. The box isn't highly critical but it is starting to get annoying having to hard reset the box and I've now gotten to the point where it's common enough that the server is on a smart plug that I can remotely turn off/on when this issue crops up. Thanks in advance for your help, tabris-diagnostics-20220821-2044.zip

Hmm.. I got updated today and now Sonarr won't start. I get a repeat of this block before it finally gives up and dies. 2021-03-09 21:08:59,097 DEBG fd 8 closed, stopped monitoring <POutputDispatcher at 22543642412560 for <Subprocess at 22543642413328 with name sonarr in state STARTING> (stdout)> 2021-03-09 21:08:59,097 DEBG fd 10 closed, stopped monitoring <POutputDispatcher at 22543642573312 for <Subprocess at 22543642413328 with name sonarr in state STARTING> (stderr)> 2021-03-09 21:08:59,097 INFO exited: sonarr (exit status 2; not expected) 2021-03-09 21:08:59,097 DEBG received SIGCHLD indicating a child quit 2021-03-09 21:09:02,101 INFO spawned: 'sonarr' with pid 68 2021-03-09 21:09:02,113 DEBG 'sonarr' stderr output: Cannot open assembly '/usr/lib/sonarr/NzbDrone.exe': No such file or directory. Forcing an update seems to have resolved this though. Might be useful for other people having the same issue.

Has anyone been able to get the external (remote) client IP address to forward to the proxied server? I've skimmed a few pages and run a search over this topic but I can't find anything on getting the client's IP address to the server. For clarity I'm running a site using IIS on Windows Server 2016, with Nginx Proxy Manager fronting the public requests. My web server only ever sees the IP Address of the docker (my unraid server), which is problematic when my application has IP Address banning implemented for security - I've had to disable it incase someone cottoned on that they could effectively use my own security against me 😐

So they do, weird I get the plugin and docker updates but I've never had an OS update notification though. I'll monitor for the next update. Request two though would still be nice

Hey all, Just thought of a potential new feature this morning - had a power cut last night, signed into unraid to start the array this morning and discovered there's a new update available. Does unraid check for updates periodically and if it does, can a new notification option be added so that it sends a notification to configured clients? (I use Pushover which has been flawless so far). As an additional option to the above, could we have an option to let unraid automatically download (but not install) the update and let us know that (as well as it being available above), that it's now ready to install? Or disable the above all together as I'm guessing some people aren't too bothered. Edit: 1 and 3 already exist as pointed out.

The last backup I have was taken while the USB device was in this state.

Hi, Yes it's booting from a USB 2.0 port. I've run check disk, I got the message that there were errors and prompted to fix them, which I confirmed to do. This takes a few seconds before returning this message: I've gone into CMD prompt, into DISKPART and run to clear the readonly attribute from the drive but it never changes the current read only state from the disk.

Hi all, Pretty sure I've seen this posted before but I couldn't find it. I've just tried to update my server to the latest version of Unraid but I received an error: plugin: run failed: /bin/bash retval: 1 Trying to update plugins errored out with either the same error or a general failure: plugin: updating: community.applications.plg Cleaning Up Old Versions plugin: downloading: https://raw.githubusercontent.com/Squidly271/community.applications/master/archive/community.applications-2020.01.18a-x86_64-1.txz ... failed (Generic error) plugin: wget: https://raw.githubusercontent.com/Squidly271/community.applications/master/archive/community.applications-2020.01.18a-x86_64-1.txz download failure (Generic error) I've then got a small notification that the USB drive was not read-write. The server was rebooted "just in case" and on loading, just before the final console login, it listed a bunch of dockers that couldn't open temp folders because of read-only. I'm a little nervous as this is the first time I've actually had anything "go wrong" with my server. The server booted up, has let me log into the WebGUI and has begun its usual parity check but the error saying USB was not read-write appeared again. As a precaution, I've downloaded a flash backup. The flash drive is only a few years old and is a Sandisk Cruzer Fit 16GB drive from Amazon. Background aside, is this indicative of a failing USB drive? Here's the disk log information: If this disk is failing, can I take the backup I've just made, get a new 32GB cruzer drive from Amazon, copy the contents of the zip over, re-licence the USB drive and carry on like nothing else happened? Thanks, S.

This is a known issue in the current build and I believe a fix is going to be released on Tuesday for the main project. When this particular docker will be updated is another story. https://github.com/influxdata/telegraf/issues/6388 Hope that helps EDIT: While I know it's got nothing to do with this docker, this is why I wish unRaid would support an official API endpoint for dashboard information. For right now scraping the dashboard in C# and parsing the DOM using HTMLAgilityPack is having to suffice.. barely

Sorry bud, I didn't even know you'd posted a response - I haven't had any notifications from the forum and only noticed when I popped on to ask a question about Grafana. I can't remember what is was that I had a problem with for this container, let me post this and I'll have a scroll back and edit this once I've remembered! -- EDIT Well I looked back and I haven't got a clue what i was on about! I do have everything setup and functioning so I would be happy to answer any specific questions you might have re the setup I use at this point.

I've literally just come back to it today, tried that and was about to post that it's worked for me before I saw your post. Had no idea if it was going to work or not but it was a shot in the dark that got the mark for me. Thank you anyway!

Hi, I have a need to access the nginx.conf file to try and fix a problem I'm having with larger header sizes with IdentityServer. Specifically in relation to: https://stackoverflow.com/a/48971393/4953847 How can I set the following values for this container? http{ ... proxy_buffer_size 128k; proxy_buffers 4 256k; proxy_busy_buffers_size 256k; large_client_header_buffers 4 16k; ... } Currently I'm able to authenticate my app but I immediately get redirected to a 502 Bad Gateway from nginx.

Stupid question numero uno.. how do you "log out" once you're in when password is used to secure? PS. This is a freaking awesome tool when I'm about with the ChromeBook.

I'd like to set up email alerts from Grafana but the settings page is read only. There's a banner that states that the settings are defined in grafana.ini or custom.ini or set as ENV variables. I've tried looking in the appdata folder for Grafana but I can't see either of the name files. I've also tried setting up a variable in the docker edit page and restarting the app but it doesn't work. How do I go about changing the settings for Grafana? Thanks,

I'm able to open it in WinRar if I can get you a particular file or two from it? tower-diagnostics-20190303-2252.zip

When configuring a proxy host, it does have an option to enable blocking common exploits, but I haven't found any documentation for what this entails.

This docker allows you to create user access lists and assign them to specific proxy hosts. From the main dashboard, click Access Lists in the menu. Create a new list and specify a username and password (up to five distinct users). Once created, go to the proxy host you want to secure, click edit to open the modal and at the bottom of the modal, select your access list from the dropdown and save. From the help text:

If you already have a web server hosting those files inside your network, then exactly where they are is fine, just point the reverse proxy at that server. If you don't have a web server running already, there are plenty of web server docker containers in Community Applications that will fit your needs. I'd suggest getting familiar with how your web server is configured and making sure it's secure before giving the outside world access to it.

Absolutely, I agree - for the majority of users the basic Nginx container suits them fine. How easy (or hard) do you think it could be to write a plugin to sit within unraid to integrate with it? I might actually have a go if I can get some time.

Without having paged through all 102 pages of this one, I do have a question. With this container acting as a proxy/gateway for all the configured services, are there any plans for (or does anyone think it might be worth having): An internal admin/control panel for the web server Ability to configure sites/services without needing to create config files or restart the container Ability to "switch off/on" a configured service within a control panel Some stat's on what services are being used (traffic by service over time, average response times etc..) I just think these could be really useful or am I wrong?

Just as a matter of closure, I managed to achieve everything I wanted to (docker apps + windows server vm serving asp.net app). It's actually been quite fun discovering that the LetsEncrypt app is handling the SSL for me - IIS on the server only knows how to deal with HTTP but with this I get full SSL on my public sub.domain.com address. So I'm happy as a clam (until someone tells me I've done something critically foolish, which is bound to happen sooner or later!)

Out of curiosity and inexperience. I eventually want to pass through a Windows VM running an ASP.NET application, but with the router forwarding all requests to the letsencrypt container I figured I should start there with something less complex first and work my way through. Also, what better web server are you referring to? The LetsEncrypt one?

Hi all, I'm having a small problem trying to get reverse proxy working for specific containers. Following SpaceInvaderOne's YT video (https://www.youtube.com/watch?v=I0lhZc25Sro) - I have Radarr, Sonarr & Heimdall working perfectly fine, but I'd like to pass through the chvb/docker-apache-php docker (named "apache"). Things I've done; I have a custom domain with subdomains CNAME'd to a no-ip ddns address (working) I have passed through all the required ports on my router (working) The LetsEncrypt container has my custom domain set and subdomains are listed (unraid,radarr,sonarr,heimdall,apache,) and it starts and successfully as per the video There wasn't a sample conf file in the LetsEncrypt container for the apache container, so I created one (apache.subdomain.conf) and copied the contents of the heimdall conf file into this file and changed: server_name apache.*; set $upstream_ApachePHP apache; proxy_pass https://$upstream_ApachePHP:8443; # Matches apache container host port 2 :8443, container port :443 Saved, and restarted the letsencrypt container. The problem I'm getting, is that while Radarr, Sonarr & Heimdall work perfectly, if I navigate to apache.mydomain.com I get a 502 Bad Gateway error from nginx. I also get the LetsEncrypt default web site when I try to access unraid.mydomain.com, which is fine as I don't need it, but I don't know why this works. If I change the apache.subdomain.conf file and only change set $upstream_ApachePHP apache; to set $upstream_ApachePHP heimdall; I get Heimdall and not apache (as I expected), so I don't know if this is an issue with the apache container or something else. If someone can help point me in the right direction that would be really great! Thanks

I know this is old, but a potential explanation for the solutions out there now is that they are simply screen-scraping. The login process is fairly simplistic using basic auth so it's not too much trouble to pull data from pages you know the structure of. I haven't personally looked at the DOM for unRaids web UI, but I'm willing to wager that if no official API exists, then this must be happening. My own take on the law of parsimony there.. I would still be very interested in an official API - I roll my own dashboard bringing together monitoring information from various restful web services and since moving to unraid, would like to continue the same.