Jump to content

Trembler

Members
  • Content Count

    16
  • Joined

  • Last visited

Community Reputation

2 Neutral

About Trembler

  • Rank
    Member

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. There still seems to be an issue with the Dockers not being able to "apply update" downloading 0 bytes from the repository. As originally reported here https://forums.unraid.net/bug-reports/stable-releases/dockers-wanting-to-update-but-dont-in-the-end-r618/?fbclid=IwAR1MYec7WcYbrpVuwywsf1-IL8gkmuQuFoqifM8cpphAPAWs-d6JeG4FFvQ Upgraded from 6.7.2 which I originally ran the user script on to resolve the update issue, was working fine, the update broke it again. After updating to 6.8 the docker update issue was still a problem thinking it was supposed to be fixed. I ran the user script again and was able to apply updates without issue. Still seems to be present as we are having other users reporting the same on our unRAID Facebook group.. Thanks!
  2. Hello from the unRAID FB group! Happy bday!
  3. I actually found the problem for some reason with 6.7rc5 it disabled perl. Once I re-enabled it, updated it, I was able to reinstall 5.8 and the plugin just fine.. I had no clue that got disabled but that was the cause.. and thanks i do see above he mentions "to fix on reboot" I will change it.
  4. Trembler

    SNMP

    Maybe I am missing something here with SNMP. This would seem to be a pretty basic feature that should be available natively. The current SNMP plugin is broken even with the available SNMP net-snmp5.8 pkg which was the workaround to get the plugin to work. Doesn't appear the plugin is being supported anymore but I'm curious why the native OS doesn't have this available? I've been monitoring a bunch of servers with Zabbix and now since the plugin is completely broken I've lost SNMP functionality.. Anyone feature request? Or fill me in on what I'm missing? Thanks!
  5. unfortunately this workaround isn't working anymore. Even with net-snmp.5.8 the SNMP plugin is failing on 6.6.7 and 6.7rc5 . Also ive noticed when rebooting the server it reverts back to the net-snmp5.7 version and you have to reinstall it each time then reload SNMP plugin.. But I guess that's beside the point now since the plugin isn't working any longer. I wish they would just build in SNMP, doesn't seem too difficult.
  6. Do you mind sharing your banner? Thanks!
  7. Excellent Job everyone the new dashboard is fantastic. Nice to get proper Temp placement and realtime utilization across all aspects on the dashboard.
  8. Is there a way on the backend to default it to use port 587? If I wanted to use the web interface it seems by default it uses port 25 and I haven't found a way to change it.. Since my ISP blocks outbound port 25 this is an issue.. Emails get delivered if i setup an SMTP relay with port 587 but i cannot use that solution going forward it was for a test... I also cannot receive any emails which I am not sure why yet but am able to telnet to my server and can see a connection on the interface so i know inbound port 25/587 is open and working. Looks like most stuff is ending up in the "queue" and just stays there.
  9. So that was the issue! Now you can see proper connection! Awesome! now the fun begins to change all my stuff over to the custom docker network which I am not even sure some of the stuff I run behind proxy can be set to a custom bridge that it must stay as a host! but at least I know what the fix is! I've actually had the custom docker network setup before but must have never fully implemented it! Now I understand about the proper dns resolution between the containers. Never realized by default the bridge didn't do it. Thank You for pointing me in the right direction!
  10. Thanks, I've already tried the network aspect of it with the same results as far as RR, single port, bond etc. The odd thing is no matter what i change I'm getting no difference in the result so it's obviously something else I haven't tried. Let me try a separate docker network for these dockers and see if that helps with DNS resolution. Although DNS resolution itself is working fine as these URLs are accessible from inside and outside with the same name but maybe something within the docker itself.
  11. Thanks for the reply. I am pretty familiar with the reverse proxy setup as I have it working with tons of other things. My friend has the exact same code setup and it's working and as a matter of fact i am routing it through orginzr so I know even on my end the code works. You can connect to the URLs just fine from the outside and inside the network as well. Its just went i try to connect these two services to each other they do not connect. wget is my way to test the connectivity and I should be getting a response with wget not a "loading" or "connecting" and then a timeout. I don't know why that is occurring! Even from other dockers to these ones they do the same. if i eliminate the reverse proxy it's the same result. I took my firewall and mapped 443 direct to the document server and still have the same issue. You are thinking a switch issue? All my other dockers are on the same switch and same port Bond0 and are working fine. I did have have some flood issues before with round Robin and the STP but I since fixed that with 802.ad and a RSTP. Still hasn't made a difference.
  12. I'm not even sure where to post this question. But here goes with the best way I can try to explain what is going on. Maybe someone knows what is the missing part of this. I've been at trying to resolve this issue for almost 2 weeks now, working on it everyday. unRAID version 6.6 Unifi USG pro4 Firewall 24 port USG switch 4 port NIC bond just using 802.ad/RSTP Myself and a buddy of mine have been trying to setup onlyoffice communityserver and documentserver with nextcloud and letsencrypt etc.. We set it up properly as his works and mine doesn't. We have identical mirrored setups as it pertains to how the containers are setup, certs, public dns , configured, routing etc.. so I know as far as the backend configuration of the dockers, things are capable of working based on our setup and his setup working. of course our hardware is very different. Here is where things get a bit interesting that I can't figure out. Currently my unifi is passing port 443 > 444 which is the letsencrypt container so all HTTPS traffic is routed through LE using /subfolders or subdomains.. I run a ton of other dockers and they all work fine.. so it does appear my firewall is routing traffic properly. Doing a tcpdump on 443 appears to confirm that as well. When I go to test nextcloud connectivity to the documentserver which is on port 4433 HTTPS (LE is using subdomain so its doing 443 > Internal IP 4433) all i get is a "loading" or "connecting" Running wget commands from nextcloud for example gets the below attachments. Nextcloud to DS > just shows connecting, no resolve like it doesn't even see the port being available. DS > nextcloud > resolves but then gets stuck on connecting and never goes any further. Also if i randomly pick any other container lets say Plex for example and try to wget to DS i get the same results so its not exclusive to just those two containers. To try and eliminate the obvious I've disabled LE Container, Mapped 443 direct to the DS docker to try and eliminate anything in the middle > removed certs from the containers (thinking wget wasn't working due to a cert issue) but exact same result.. no matter what I try to disable or remove from the equation I am seeing the exact same results.. My firewall logs are showing sucessful ACK of the packets but I'm also seeng IPV4 Martian errors from 172.17.0.10 > my Public IP That is the Docker Bridged network for the Documentserver Container. We are both baffled by what is occurring here and i'm not sure wherelese to look to try and get some answers.. It seems as if something is blocking something but I have no idea what (common sense would say the firewall) but everything else is being routed the exact same way through LE/443 and works fine.. Im talking another 10 or so dockers.. My next step was to try a DMZ setup or just pickup a cheap router/firewall toss it in there and see if the results are the same.. What on unRAID could possibly cause something like this? Bad Docker network etc? even if everything else is working fine? I've probably removed/reinstalled these dockers 15 times, same results every time! all the URLs work fine with the proper certs on HTTPS. I will try to provide some more logging but I am really not seeing anything. Thanks for any direction provided.
  13. Thank you, makes total sense! It's confusing. I like the thought of not being able to select isolated cores in the containers. So basically if i isolate cores i can use them just for the VM and nothing else will touch them?
  14. There definitely seems to be an issue with CPU isolation and Dockers.. Did a bunch of testing with Plex, Nzbget, Sonarr, Radarr etc and it seems when you isolate multiple cores, for example I Isolated cores 5/11 which is my pair I assigned those isolated CPUs to nzbget, sonarr,radarr and you can only see core 5 being utilized, not 11.. It doesn't make much sense. As soon as I disable isolation on lets say core 11, I can then see usage on that core from those dockers.. At this point I've completely turned off CPU isolation and it seems all my CPU pinning is working properly but id like to isolate certain cores so the OS doesn't utilize them at all, just certain dockers.. Not sure if this is a bug or a misunderstanding on how it's supposed to work.