BCinBC

Is there any documentation on how to configure the Watchdog timer? I have searched the forums and looked briefly through the system options. I'm not sure I want to enable in the command line if it's supported somewhere else.

Thanks for that! Now I know why it got "fixed!" I have had CA Auto Update running for ages, but once I did the reboot maybe the the fix kicked in? I've been up about 6 days now and the issue is certainly gone now.

I found, that recently, all kinds of dockers were getting Zero Byte downloads and non-removable updates. i.e. You'd update it, it wouldn't really do anything but restart the container. Then when you checked it again, it wanted another upgrade. I rebooted the server and this went away. If it happens again I'll definitely file a bug report. It had been up for maybe 45+ days?

I tried to install this on my server so that I could remotely access various dockers. However when I activated ZT the main IP address of my management interface disappeared. The only way I could access the server's management interface and all of the various devices on it was only by using ZeroTier even within my own network. Is it supposed to do this? I would like to be able to connect to the server within my own network as well as be able to remotely access it with say an Android Phone.

I found that changing the docker networking mode from Host to Bridge and manually mapping all of the ports needed, It's been rock solid ever since. As soon as I advanced to the new version that had the new network code it was unstable. This seems to affect only a very small group of people but I see it hit the forum every few weeks. From My plex forum post: https://forums.plex.tv/t/plex-server-not-worked-since-1-13-4-5271-200287a06/312196/12?u=bobertc

Custom: br0 Needed for Unifi, & PiHole. It also seemed to kill Splunk, which is just a Bridge connection. So how does one create a new External IP for a docker with 6.6?

Did anyone else have their bridge configuration disappear? I had a few dockers with fixed IP addresses and they suddenly lost Bridge:0. I'm assuming that this happened with the new bridge utils I saw mentioned in the package updates. I had to undo and redo the settings after backing down to 6.5.3. Love the new UI look though. I posted a bug report.

It also happened to me on an Ubuntu install on an ESXi server, completely outside of docker. I've been running Plex in my environment for 4-5 years without issue on all kinds of platforms, mostly Linux. This version, I cannot go to as the server cannot phone home to Plex properly. I'll try it on Bare metal if someone really would like me to. It may be "environmental" too. I was thinking of trying it on a second network under pfSense instead of the Ubiquiti ER-X it is now.

https://forums.plex.tv/t/plex-server-only-online-for-30-seconds-every-7-minutes-or-so/293805 Sorry link isn't working on my phone client. That's what I opened with Plex.

I tried it, it didn't fix anything for me. I went back to 1.13.2, all still working there.

I do understand about the version, I'm not trying to bypass anyone's eligibility (I have a lifetime PlexPass) I'm just passing along what I was recommended by a Plex support person, they said to fix this issue, revert to 1.13.2 https://forums.plex.tv/t/plex-server-only-online-for-30-seconds-every-7-minutes-or-so/293805/35?u=bobertc ChuckPATeam Member Regarding the “Lost Internet Connectivity” issue we’re all seeing. 1.13.5 introduced the ability to bind to an adapter. That works great. Unforseen side effect is it disturbed MyPlex. Engineering is working on that now and , AFAIK, will be in next update. BUT, i’m deep in NAS work so don’t have the absolute latest info on it.

@saarg You are correct, mea culpa. I did this on Plex's docker image instead. I needed a fix after being down after 3 weeks, I've tried for weeks to find a fix. Plex has messed up the My Plex code and is causing a lot of grief for specific individuals. Reverting to 1.13.2 is the recommended fix at the moment for those that have Plex stability issues. I actually did this on a different docker Image. In my haste of the excitement of finding a fix, I didn't realize that I had switched images to: "plexinc/pms-docker:1.13.2.5154-fd05be322" That pulled correctly for me, but I would definitely have preferred to use linuxserver.io's version instead.I tried setting the VERSION variable to a previous version but I was unsuccessful to find a working one. If you followed my stupid advice, you'll want to go back into your "Add docker" remove my advice and put in the "repository" field back to just "linuxserver/plex" that should pull the correct image again, but not fix your version number. @saarg If you actually have the Linuxserver.io's 1.13.2.5154 image number, I'd gladly try it in the VERSION field.

Don't do this, this was done on a different Plex image. If you followed my stupid advice, you'll want to go back into your "Add docker" remove my advice and put in the "repository" field as just "linuxserver/plex" that should pull the correct image again, but not fix your version number, which is why you're looking for a fix in the first place. Add ":1.13.2.5154-fd05be322" After 'linuxserver/plex' So it looks like this 'linuxserver/plex:1.13.2.5154-fd05be322' then apply / done. I then removed the container and removed all of the files, tick the checkbox when you remove it (This does not remove the data. Remember that docker stores your user data in the appdata area). This is just to make sure there were no residuals. Then re-install the docker from your own templates (it should now have saved the value you added above) It's been running for about 8 hours with no connectivity issues. Also make sure that you are Not automatically updating your docker. I noticed that once I forced it to this version it no longer says it has an update.

@aptalca will this eliminate any network throttling and have the system assume these are all local connections?

Aaah, OK, because my problem isn't' getting to my server remotely... it appears to be my server "getting out" so this may make that part happy.

@aptalca The issue is with internal and external connectivity. But this is a great piece of info!

That's a neat trick. Look in this directory for previous reports: /mnt/user/appdata/PlexMediaServer/Library/Application Support/Plex Media Server/Crash Reports However, mine's pretty much empty because I've re-installed plex about 13 times with 3 different distros and also blew away the docker file and app file.

@subagon Plex specified 1.13.2 is needed to go back far enough before this networking issue. There's a new option that recently appeared under the Server -> Networking option where you can tell Plex which NIC to use. That code doesn't work properly yet it would seem on docker and VM style containers.

I know that you can specify a build name that the dev has done usually like plexpass, beta, developer etc. But I didn't think it was that trivial to specify a Plex (or other) version number. Thanks, I'll check it out when I get home. Answers are brief as I'm on the train home. I Need 1.13.2, apparently.

If it comes up for a few seconds every few minutes. Plex is aware there is a problem, not sure when it will be fixed. See my previous post on this issue. I have a link to the Plex forum with my post on it. The fix, for now? Go back to the previous version. (and no, I don't know how to do that. I wish I could too)

Oh, so if you want to change networking and adding bridges, if I recall, you need to spin down your array before you can change the advanced network settings to allow you to add virtual NICs and bridges. I spent so much time over the past 10 days with this, I've forgotten what I did. I've just turned off Plex until they come out with a new relese. I've thrown in the towel and am using unRAID with my KODI boxes.

Docker natively has a bunch of networking, subnets, etc. that you're not going to be able to remove. Plex tried to make the code that would allow you to specify which network interface to use. While you *CAN* specify that, the code underlying this detection method broke in the last release. They're working on it, and so I wait, I can't get any Plex server to run in my environment under Docker or Linux, ATM. A quick test is to launch a web client and try to connect to the server. Mine literally is available for 30 seconds every 5 minutes or so. That's the acid test I found. It won't show up initially for a few minutes too.

I'm going to add this here, according to @ChuckPA at Plex.tv https://forums.plex.tv/t/plex-server-only-online-for-30-seconds-every-7-minutes-or-so/293805/30?u=bobertc This completely messes up the docker experience, I have a sneaking suspicion we're going to see a lot more complaints about Plex and docker implementations until this gets fixed. I hope I saved someone some time, I've spent way too much time in the past week on this.

I've been bashing my head against the wall for the past week as I too have had my Plex docker container go offline, it stay up for about 30 seconds every 5-7 minutes. I've got a discussion thread on the Plex forums about it. At first I thought it was a network DNS rebind issue (wasted 4 days of my long weekend on that). Plex recently has changed how the networking is done, there's also a new Advanced server feature that lets you select which network interface to use that apparently just got added. Since last week, no container I've used, be it, linuxserver.io, Binhex's plex or his plexpass. All three have the same issue. I even fired up my old previous unRAID box, Chimera. (I saw this server name earlier in this thread. The current one is named Bellerophon.) I also spun up a full build of Ubuntu 16.04 LTS on my ESXi box and it too (which made me think it was my network) had this problem. I haven't revisited it yet and tried the new watch network interface, but the VM only had one interface which was directly bound to my local network. Reading the logs as soon as they say "Aug 07, 2018 10:05:02.800 [0x14d5c5bff700] DEBUG - MyPlex: We appear to have regained Internet connectivity." The server vanishes and won't appear for another 30 seconds in a few more minutes. https://forums.plex.tv/t/plex-server-only-online-for-30-seconds-every-7-minutes-or-so/293805/7 Just to be clear, this is on my internal network, the issues of trying to watch it remotely are at the mercy of the server being up or not. I've tried two of the four interfaces that appear, I've tried it in both host and bridged mode too assigning it an external new IP address, same results. So before I try firing up a windows box to use plex with (shudder) please, anyone have any ideas?