Server Version#: 1.21.3.4021
Player Version#: Various, such as Web 4.52.1 and iOS 7.13
Just recently (I believe it was right after I updated from 1.21.1 or 1.21. 2 to 1.21.3.4021), I’ve noticed this symptom. I’m assuming it’s PMS crash, but let me know if there’s some way to be sure.
I’ll go to watch some content in some client (Web, iOS, Android TV, etc.) and my server will act like it’s offline. I can ping the QNAP server from these devices, and QNAP’s App Center reports PMS as running. I can STOP and START the app from the QNAP App Center, and all the clients will show the server come online within moments.
This happens frequently now - every few hours, sometimes within minutes. I can watch content when it works. This problem has been going on for a couple days now, but before then I’ve not had any issues like this for many, many years with the same setup.
I can get the server logs after I restart the server, but what can I look for in those logs to help understand what’s going on? And how can I tell which log file(s) are relevant (e.g. I think a log gets created the moment I restart the server)?
The server runs fine until I try to use it from my most used client: the Android TV client on the SHIELD. As soon as I start that client (since the last few days) it will fail to load content, and at that point no other client will load the content (until I stop and start the server).
Android TV client is 8.12.4.22902 updated on 2/8/2021.
DEBUG log files would really help to figure out what’s happening so it can be fixed.
I know of one Android app problem they’re working on.
I also know of a possible server problem but have only one possible instance of it in log files.
If folks can have DEBUG logging on, VERBOSE logging off, and capture / recreate and post those logs, it will go a long way to helping identify and fix whatever is happening.
I have an idea but I need the smoking for Engineering.
Sorry, I see too much personal info in DEBUG log files. Otherwise I would have posted them with my first post. Not to mention I never got an answer to my question in my first post re: which log files to look at (so that I can help you find that smoking gun).
I rolled back to PlexMediaServer-1.21.1.3876-3c3adfcb4-x86_64.qpkg by just manually installing it (the same way I upgrade).
It reproduced immediately there.
Then I rolled back to PlexMediaServer-1.21.0.3616-d87012962-x86_64.qpkg. Same thing. And I know I wasn’t having this problem back in December and January when I ran these versions. I’m not sure how Pablinx was able to avoid this by downgrading their server - I’d like to know what version they went to that avoided the problem.
So I believe this has something to do with the new Android client release 8.12.4.22902. This makes it very hard to workaround this problem because I have a lot of these Android clients installed in multiple locations.
I see this in the “Plex Media Server.1.log”, referencing IPs running that Android client. Not sure if it’s related.
Feb 14, 2021 10:58:12.736 [0x7f4229144700] DEBUG - NetworkServiceBrowser: Parsing SSDP schema for http://192.168.1.16:9080
Feb 14, 2021 10:58:12.736 [0x7f4229144700] DEBUG - HTTP requesting GET http://192.168.1.16:9080
Feb 14, 2021 10:58:12.743 [0x7f4229144700] DEBUG - HTTP 200 response from GET http://192.168.1.16:9080
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - XML: Entity: line 1:
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - XML: parser
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - XML: error :
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - XML: Start tag expected, '<' not found
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - XML: status=ok
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - XML: ^
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - Error parsing content.
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - Error parsing XML: Error parsing file.
Feb 14, 2021 10:58:12.743 [0x7f4229144700] ERROR - SSDP: Error parsing device schema for http://192.168.1.16:9080
Sorry to be a bit pushy but we can’t isolate and fix without seeing the whole sequence. “ERROR” snippets are grossly inadequate. Engineering needs to see what PMS was attempting to do when the ERROR happened.
As for those SSDP? It’s an annoyance which we’re trying to get suppressed. It’s otherwise harmless. PMS is too vocal about reporting bad replies during device discovery.
Once I see the personal info removed from what you’re asking for, I’ll be happy to post them. What I’ve seen with just a quick look at one of dozens of files in the zip are account names and folder names (which for me includes locations, real names, etc).
Hopefully this one is common enough for engineering to see it from their own testing. Or someone finds a workaround. I have not tried sideloading older versions of the Android client yet (I don’t have them and haven’t gone looking for those APKs).
It’s the first time in about 5 years of heavy use that I’ve hit an issue that’s completely preventing me from using Plex.
for Engineering.
