PMS is somehow unable to be configured, after 2-3 days of running

Server Version#: now 1.18.2.2015, but version before (including last 2 betas and stable) had the same problem.

Player Version#: latest Plex.app MacOSX and Windows7, as well as Plex Web GUI

I’m on: Synology NAS DS1815+ 16GB RAM, DSM 6.2.2-24922 Update 3

When running the PMS longer than, say 3 days, I cannot connect to the settings anymore. (Which is everything under the SETTINGS menu-entry)

I’m getting a message like: unable to connect… or settings not available.

The PMS itself is working. I can watch movies or listen to music. When clicking on STATUS Alerts, I can see, its scanning a library.

When I stop PMS via Paketmanager and start again, everything works as expected. I can configure the server and change settings. After 2 or 3 days, no configuration is possible anymore, I’m not able to access those menu-entries.

When I manually try to fix a match, I only get this:


(The “Loading…” will never vanish and I cannot choose an Agent)

When I’m already in the WebGui on the MANAGE console page, it still shows ongoing activity, but when I reload the page by hand, no connection is possible anymore.

Is this a known issue?

When it sits in “Loading” like that, make sure the Synology has:

  1. Proper connectivity to the Internet
  2. DNS entries are correct for your ISP and/or pointing to Google
  3. The “Default Gateway” is set in Control Panel - Network
  4. Uninstall the package itself from Package Center (your data is safe)
  5. Restart DSM
  6. Re-install the package SPK.
  7. Wait 45-60 seconds before opening PMS

If this is all correct, we can do it again but save your existing content while we diagnose.

  1. Stop PMS
  2. Control Panel - Shared Folders - Plex share
  3. EDIT the Plex share
  4. Permissions tab
  5. Give your Synology username permission to R/W the share.

Now open FileStation. Navigate to Plex share.

  1. Rename Library -> Library.old
  2. Start PMS
  3. Wait 2 minutes for it to perform first-run startup
  4. Open http://ip.addr.of.syno:32400/web in your browser.

Thanks for the detailled description!

I already have done/checked all of this myself the last weeks, it did not help.

Network settings are good, I say this, because the Plex configuration pages are working the first 3 days, then Plex is somekind “overloaded”.

I guess, its the amount of data, Plex has to read and index.
it now takes over 3weeks to index 60%-70% of the music. (*)

BTW:
When I start → Troubleshooting → Optimize Database, Plex settings are unrechable too. If optimizing is done, everything works again.

Its not the network settings…

(*) The agent settings:

Music? Oh no! :scream: lol

It’s never good to throw a lot of anything at the server at once when running on a Synology. The CPU is not very strong.

Plex doesn’t have a limit as to how many folders you can give it to monitor so, if you have it broken up a little bit, you can use that to your advantage.

Instead of giving it /volume1/music
you can:

  1. Start with /volume1/music/rock
  2. When finished, add /volume1/music/folk
  3. etc

However you have it curated into sub-folders is just fine.
You will be MUCH happier on a Synology this way

Synology is a DS1815+ with an Intel Atom C2538 Quad Core, which is more powerful than my desktop computer :wink:

So you recommend to add 27 single libraries?
because I have set my music-folder this way:
music/0
music/a
music/b
music/c

music/z

EDIT:

So I am the only one who has this problem?

The DB com.plexapp.plugins.library.db is only about 2GB in size. Optimizing is should not be a big thing. Maybe reorganizing the trillions of thumbnails…but thats not CPU, its diskspeed.

So why is only the configuration pages inaccessible? All the rest is (thank god) working.

I have that model too. I know it well.

Let’s get it back to being “your” syno. It will get lost when it gets all out of whack on a synology.

  1. stop PMS
  2. Open FileStation
  3. Go into the Plex share and drill down into “Plex Media Server”
  4. Rename “References.xml” to “Preferences.xml.old”
  5. Start PMS

If you don’t see the Plex share -> Control Panel - Shared Folders - Plex - EDIT - Permissions tab. Add your username.

Now, once you’ve renamed the preferences file, we’re going to have Plex.tv forget the Synology for now.

Open app.plex.tv/desktop

Settings - Authorized Devices - “Servers”

Remove the Synology fo now.

Next we start up the DiskStation and open it directly by IP from your LAN (you must be local)

http://ip.addr.of.syno:32400/web

  1. Be greeted by the “Got It”
  2. Sign in
  3. Start the Setup wizard EXCEPT Do not create any new Library sections. we’re just resetting ownership but it doesn’t know better yet.
  4. Continue through the wizard until you get to the Dashboard.
  5. Your Libraries will be there waiting
  6. Check the rest of your preferences (some do get reset)

Let me know if this wasn’t enough

Thank you @ChuckPa ,

I did as you described, but I am stuck with

Media server not found.

The server you’re trying to access doesn’t seem to exist.

Those are the steps I did:

Stopped PMS:

Renamed Preferences.xml:

Started PMS:

Signed in:

Selected PMS to be DEauthorized:

I was sure:

Tried to start up directly from LAN (I’am local):

Now this:

What now?

anyone?

I don’t want to sound impatience, but after following the exact steps the support told me, I’m sitting on an unusable PMS.

I don’t want to test out different posts here and aggravate the situation,
so what are the next steps to get the server up and running?

(I guess its up and running, but how do I connect correctly, if not from local IP, as told?)

Sorry. I’m the sole support here now. The mainstay Ninja, Trumpy81, has moved on .
What is the IP address of the Syno? (LAN address)

Is it 10.x.x.x, 192.168.x.x, or 172.16.x.x -> 172.31.x.x ? If not, you need SSH tunnel to it because it’s not on a Private Network as defined in RFC-1918 (see WikiPedia)

Its a local IP, 192.168.168.111

EDIT:
somehow, and by chance, I started the Plex.App on my iPhone to find some new bugs via Testflight :slight_smile: and a message popped up, telling me, that there is a new and homeless server waiting to be claimed.

So I clicked the “yes, please claim this server” which could only be the server, I couldnt reach via local IP as in the screenshots above:

(Unfortunately only german language screenshots)


I then could continue the steps you told me in a desktop browser:




So, after claiming the server via the iPhone, I now can connect to the local IP 192.168.168.111:32400/web, but unfortunately, the server itself is not reachable:


(On the left side, all the pinned libraries with (!) are from the server (192.168.168.111) which is not accessible.

What can I do now?

EDIT2:
Ok,

HTTPS instead of HTTP was the deal.

I guess, the Plex network settings were set to default, when renaming the Preferences.xml in the first place. I did not took that into consideration…

I guess so, because I never had IPV6 activated, but now its activated:

Security is set tp “Prefered”, and I was under the impression (told in other threads by another supporter, AFAIK), that HTTP was the correct way to connect and plex will “secure” automatcly, when possible. But hey, another day, another solution :slight_smile:

It now works again, and I will report if this was or was not “enough” for the server not to be inaccessible after 2-3 days of scanning…

I will close now with a nice quote by Bart Simpson: “God, we paid for all this ourselves, so thanks for nothing.” :wink:

Guess what… it was NOT enough… same as before:

When the scheduled task of Database optimiting is active, server-settings are inaccessible.

When the process of Database optimizing is finished, server-settings are accessible again:

So now, it should be clear, that no user-setting is responsible, but
either:
Disk-Performance? (It’s a RAID6, 32TB, 8x5400RPM WD Drives)

aka

Slow SQL access?

Log shows a lot like these:

Nov 14, 2019 17:24:08.719 [0x7f403511f700] WARN - Held transaction for too long (../Library/DatabaseFixups.cpp:207): 0.990000 seconds
Nov 14, 2019 17:24:25.586 [0x7f404d71b700] ERROR - EventSource: Retrying in 600 seconds.
Nov 14, 2019 17:25:28.308 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:17): 54.930000 seconds
Nov 14, 2019 17:25:53.355 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:27): 20.240000 seconds
Nov 14, 2019 17:25:59.854 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:38): 5.720000 seconds
Nov 14, 2019 17:26:05.020 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:48): 3.240000 seconds
Nov 14, 2019 17:34:25.639 [0x7f404d71b700] ERROR - EventSource: Retrying in 600 seconds. 

(That was at the same time, the Database was optimizing)

Guess, I’ll try to put the Database on a RAMDISK to speed up some things..

See the error?

Nov 14, 2019 17:24:25.586 [0x7f404d71b700] ERROR - EventSource: Retrying in 600 seconds.
Nov 14, 2019 17:25:28.308 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:17): 54.930000 seconds
Nov 14, 2019 17:25:53.355 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:27): 20.240000 seconds
Nov 14, 2019 17:25:59.854 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:38): 5.720000 seconds
Nov 14, 2019 17:26:05.020 [0x7f403511f700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:48): 3.240000 seconds
Nov 14, 2019 17:34:25.639 [0x7f404d71b700] ERROR - EventSource: Retrying in 600 seconds. 

Event Source 600 Seconds?

The machine has lost connectivity to the Internet.

The WARN messages tell me the machine is thrashing badly

  1. How much RAM?
  2. What else is it doing that that time? Chapter image generation?

To my understanding, it says “retrying in 600 secs”. its not taking 600 secs. But the held transactions are taking endlessly long…

I don’t have a clue, what “Event source” means…

Your questions:

16GB RAM.

The machine has constant internet connectivity.

My line is 100MBit Down-/10Mbit upload.

As you can see in my screenshots from above (Agents settings), Plex should only fetch local information from tags (and maybe some less from last.fm, but not very much).

Also, why should internet connectivity matter, when the database is optimizing…

Nothing CPU intensive is happening at the same time.

There is an online backup running right now and I can see data is transfered without discontinuity. (Online Monitor from my router)

A ping at the same time, when Plex is “hanging”:

:> ping heise.de
PING heise.de (193.99.144.80) 56(84) bytes of data.
64 bytes from redirector.heise.de (193.99.144.80): icmp_seq=1 ttl=247 time=24.1 ms
64 bytes from redirector.heise.de (193.99.144.80): icmp_seq=2 ttl=247 time=21.4 ms

So thats not the right trail.

The correct trail would be (I guess):
Slow database (write) access.
Or diskspeed/performance.
Plex is somehow loosing connectivity to the webinterface (timeout) or the plex-client.

Correct. It will try to connect in 10 minutes from that time mark.

PMS requires constant connection to the internet unless initially configured to run “quasi” without it.

All PlexPass related features require constant connectivity…

If the server cannot authenticate against Plex.tv, it will not know who owns it or be able to validate what it thinks it knows.

As I said, there is a constant internet connection.

Why should the connection only break away, when the database is optimizing?

Makes no sense to me, but I’m open to any technical explanation :slight_smile:

Also my other PMS on a Win7 machine, which is in the same network (and uses the same internet connection), is working w/o any problem (when the PMS on the syno is not)

So, to get to the point:
How can I track down the problem to get a non-stop-running PMS?

Do you know, what options/tools I can use to identify the cause of the “timeouts”?

Here is the logfile, when starting the Plex package.

No scanning or optimizing has been starting, but interpreting those lines seems there is a serious problem with the database access/queries. Whatever that could be, I don’t know.
No movies/music was played.

What could be the reason, PMS is so extremely slow?

Nov 14, 2019 23:02:16.733 [0x7fa050ae1700] INFO - Plex Media Server v1.18.2.2029-36236cc4c - Synology DS1815+ x86_64 - build: linux-x86_64 synology - GMT 01:00
Nov 14, 2019 23:02:16.733 [0x7fa050ae1700] INFO - Linux version: DSM 6.2.2.24922-4, language: en-US
Nov 14, 2019 23:02:16.733 [0x7fa050ae1700] INFO - Processor         Intel(R) Atom(TM) CPU  C2538  @ 2.40GHz
Nov 14, 2019 23:02:16.733 [0x7fa050ae1700] INFO - /var/packages/Plex Media Server/target/Plex Media Server
Nov 14, 2019 23:02:19.129 [0x7fa05b18d740] INFO - OCSP: Successfully retrieved response from cache.
Nov 14, 2019 23:02:29.320 [0x7fa03a583700] ERROR - Error issuing curl_easy_perform(handle): 7
Nov 14, 2019 23:02:29.320 [0x7fa03a583700] WARN - HTTP error requesting POST http://127.0.0.1:32600/devices/discover (0, No error) (Failed to connect to 127.0.0.1 port 32600: Connection refused)
Nov 14, 2019 23:02:38.985 [0x7fa01bfff700] WARN - Held transaction for too long (../Library/MetadataCollection.cpp:237): 2.320000 seconds
Nov 14, 2019 23:02:39.185 [0x7fa038dc4700] INFO - Sync: downloaded 0 sync list(s) with 0 sync items(s): 0 new, 0 updated, 0 deleted
Nov 14, 2019 23:02:39.196 [0x7fa019fc5700] WARN - Sync: local sync directory "/volume1/Plex/Library/Application Support/Plex Media Server/Cache/Transcode/Sync+" does not exist
Nov 14, 2019 23:02:39.376 [0x7fa01bfff700] WARN - Held transaction for too long (../Library/MetadataCollection.cpp:237): 0.130000 seconds
Nov 14, 2019 23:02:39.580 [0x7fa0390b2700] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
Nov 14, 2019 23:02:39.588 [0x7fa01bfff700] WARN - Held transaction for too long (../Library/MetadataCollection.cpp:237): 0.170000 seconds
Nov 14, 2019 23:02:39.681 [0x7fa0390b2700] INFO - Network Service: Abandoning browse socket, it was closed.
Nov 14, 2019 23:02:40.034 [0x7fa03a871700] ERROR - getaddrinfo(192-168-168-111.abcdefghijklmnopqrstuvwxyz012345.plex.direct) failed: -5
Nov 14, 2019 23:02:40.352 [0x7fa01bfff700] ERROR - getaddrinfo(192-168-168-111.abcdefghijklmnopqrstuvwxyz012345.plex.direct) failed: -5
Nov 14, 2019 23:02:41.758 [0x7fa01ba23700] INFO - AutoUpdate: no updates available
Nov 14, 2019 23:02:44.046 [0x7fa03a871700] ERROR - getaddrinfo(192-168-168-111.abcdefghijklmnopqrstuvwxyz012345.plex.direct) failed: -5
Nov 14, 2019 23:02:50.165 [0x7fa03a871700] ERROR - EventSource: Retrying in 15 seconds.
Nov 14, 2019 23:02:50.901 [0x7fa019fc5700] ERROR - Unknown metadata type: 
Nov 14, 2019 23:02:51.189 [0x7fa01ae6b700] ERROR - Unknown metadata type: 
Nov 14, 2019 23:02:51.564 [0x7fa019fc5700] ERROR - Unknown metadata type: 
Nov 14, 2019 23:02:52.772 [0x7fa03a583700] ERROR - Unknown metadata type: 
Nov 14, 2019 23:02:54.244 [0x7fa03a583700] WARN - SLOW QUERY: It took 1210.000000 ms to retrieve 0 items.
Nov 14, 2019 23:02:55.442 [0x7fa03a583700] WARN - SLOW QUERY: It took 1120.000000 ms to retrieve 0 items.
Nov 14, 2019 23:03:05.170 [0x7fa03a871700] ERROR - EventSource: Retrying in 30 seconds.
Nov 14, 2019 23:03:07.012 [0x7fa0393a0700] INFO - LibraryUpdateManager path watching is disabled
Nov 14, 2019 23:03:10.309 [0x7fa03a583700] WARN - SLOW QUERY: It took 1170.000000 ms to retrieve 0 items.
Nov 14, 2019 23:03:35.173 [0x7fa03a871700] ERROR - EventSource: Retrying in 60 seconds.
Nov 14, 2019 23:04:35.178 [0x7fa03a871700] ERROR - EventSource: Retrying in 120 seconds.
Nov 14, 2019 23:06:35.182 [0x7fa03ab5f700] ERROR - EventSource: Retrying in 240 seconds.
Nov 14, 2019 23:10:35.346 [0x7fa03a871700] ERROR - EventSource: Retrying in 480 seconds.
Nov 14, 2019 23:18:35.371 [0x7fa03a871700] ERROR - EventSource: Retrying in 600 seconds.

This log (an hour earlier) shows mostly the same timeouts:

Nov 14, 2019 22:38:32.016 [0x7fa720619700] INFO - Plex Media Server v1.18.2.2029-36236cc4c - Synology DS1815+ x86_64 - build: linux-x86_64 synology - GMT 01:00
Nov 14, 2019 22:38:32.017 [0x7fa720619700] INFO - Linux version: DSM 6.2.2.24922-4, language: en-US
Nov 14, 2019 22:38:32.017 [0x7fa720619700] INFO - Processor         Intel(R) Atom(TM) CPU  C2538  @ 2.40GHz
Nov 14, 2019 22:38:32.017 [0x7fa720619700] INFO - /var/packages/Plex Media Server/target/Plex Media Server
Nov 14, 2019 22:38:34.287 [0x7fa72acc5740] INFO - OCSP: Successfully retrieved response from cache.
Nov 14, 2019 22:38:44.818 [0x7fa716329700] ERROR - Error issuing curl_easy_perform(handle): 7
Nov 14, 2019 22:38:44.818 [0x7fa716329700] WARN - HTTP error requesting POST http://127.0.0.1:32600/devices/discover (0, No error) (Failed to connect to 127.0.0.1 port 32600: Connection refused)
Nov 14, 2019 22:38:54.007 [0x7fa71458e700] WARN - Held transaction for too long (../Library/MetadataCollection.cpp:237): 0.780000 seconds
Nov 14, 2019 22:38:54.835 [0x7fa714b6a700] INFO - Sync: downloaded 0 sync list(s) with 0 sync items(s): 0 new, 0 updated, 0 deleted
Nov 14, 2019 22:38:54.835 [0x7fa6e5fc5700] WARN - Sync: local sync directory "/volume1/Plex/Library/Application Support/Plex Media Server/Cache/Transcode/Sync+" does not exist
Nov 14, 2019 22:38:55.204 [0x7fa715146700] ERROR - Network Service: Error in browser handle read: 125 (Operation canceled) socket=-1
Nov 14, 2019 22:38:55.304 [0x7fa715146700] INFO - Network Service: Abandoning browse socket, it was closed.
Nov 14, 2019 22:38:55.740 [0x7fa71458e700] ERROR - getaddrinfo(192-168-168-111.abcdefghijklmnopqrstuvwxyz012345.plex.direct) failed: -5
Nov 14, 2019 22:38:55.990 [0x7fa716617700] ERROR - getaddrinfo(192-168-168-111.abcdefghijklmnopqrstuvwxyz012345.plex.direct) failed: -5
Nov 14, 2019 22:39:00.207 [0x7fa716905700] ERROR - getaddrinfo(192-168-168-111.abcdefghijklmnopqrstuvwxyz012345.plex.direct) failed: -5
Nov 14, 2019 22:39:05.102 [0x7fa6e7159700] INFO - AutoUpdate: no updates available
Nov 14, 2019 22:39:09.487 [0x7fa716905700] ERROR - EventSource: Retrying in 15 seconds.
Nov 14, 2019 22:39:10.079 [0x7fa6e5fc5700] ERROR - Unknown metadata type: 
Nov 14, 2019 22:39:10.903 [0x7fa6e6e6b700] ERROR - Unknown metadata type: 
Nov 14, 2019 22:39:11.114 [0x7fa6e5fc5700] ERROR - Unknown metadata type: 
Nov 14, 2019 22:39:12.629 [0x7fa717d47700] ERROR - Unknown metadata type: 
Nov 14, 2019 22:39:13.967 [0x7fa717d47700] WARN - SLOW QUERY: It took 1090.000000 ms to retrieve 0 items.
Nov 14, 2019 22:39:15.116 [0x7fa717d47700] WARN - SLOW QUERY: It took 1090.000000 ms to retrieve 0 items.
Nov 14, 2019 22:39:19.651 [0x7fa714e58700] INFO - LibraryUpdateManager path watching is disabled
Nov 14, 2019 22:39:24.491 [0x7fa716617700] ERROR - EventSource: Retrying in 30 seconds.
Nov 14, 2019 22:39:27.667 [0x7fa717d47700] WARN - SLOW QUERY: It took 1070.000000 ms to retrieve 0 items.
Nov 14, 2019 22:39:54.494 [0x7fa716905700] ERROR - EventSource: Retrying in 60 seconds.
Nov 14, 2019 22:40:54.498 [0x7fa716617700] ERROR - EventSource: Retrying in 120 seconds.
Nov 14, 2019 22:42:54.517 [0x7fa716617700] ERROR - EventSource: Retrying in 240 seconds.
Nov 14, 2019 22:43:53.957 [0x7fa6e7159700] WARN - Held transaction for too long (../Library/DatabaseFixups.cpp:207): 0.750000 seconds
Nov 14, 2019 22:44:48.899 [0x7fa6e7159700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:17): 53.580000 seconds
Nov 14, 2019 22:45:10.707 [0x7fa6e7159700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:27): 19.910000 seconds
Nov 14, 2019 22:45:16.451 [0x7fa6e7159700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:38): 5.660000 seconds
Nov 14, 2019 22:45:19.744 [0x7fa6e7159700] WARN - Held transaction for too long (../Library/FullTextSearch.cpp:48): 3.260000 seconds
Nov 14, 2019 22:46:54.687 [0x7fa716905700] ERROR - EventSource: Retrying in 480 seconds.
Nov 14, 2019 22:48:15.023 [0x7fa6e7a23700] ERROR - [PlexRelay] kex protocol error: type 7 seq 11
Nov 14, 2019 22:48:15.212 [0x7fa714b6a700] INFO - [PlexRelay] Allocated port 29195 for remote forward to 127.0.0.1:32401
Nov 14, 2019 22:52:01.437 [0x7fa69ffff700] WARN - SLOW QUERY: It took 310.000000 ms to retrieve 6 items.
Nov 14, 2019 22:54:54.852 [0x7fa716905700] ERROR - EventSource: Retrying in 600 seconds.

To answer your questions:

  1. Responsiveness is not the problem. The configuration pages are NOT ACCESSIBLE while the database is optimizing (and are not accessible, after 2-3 days of scanning the music folders).
    The PMS is responding great (via any plex.app or webgui).

  2. Database needs about 10 mins to optimize. I have done this about 10 times the last days. The database should not be fragmentet anymore :slight_smile:

  3. I’ll upload the log-zip, as soon I have checked the content.

How much music did you add (number of tracks)?

Music is PAINFUL if you give it everything all on at once.
Several days to scan on a Synology is not unrealistic (sadly)

The new music scanner does:

  1. Fingerprint the first track
  2. Fingerprint the last track
  3. Attempt to match using given names and fingerprints.
  4. If no match obtained, Fingerprint all tracks
  5. Attempt match again.

It will do this for every album in the queue.

Sadly I know, Plex is not the first and not the only MediaSever I tried.

Thats a good question, I have no idea, I guess, less than 700k soundfiles/music.

The Syno scans since 3 weeks. Thats a month, because I have to stop the scanner every few days and start new.

I set up the scanner to only recognize local and embedded tags, so it won’t download tons of gigabytes from last.fm and fanart.tv, but that seems to be the problem.

I have a backup server with the same amount of soundfiles/music (mirrored), but the agent is configured to get data from online sources, and its working great. (initial scan took less than a week).

I’m curious about the new music scanner.

I tested MusicBrainz Picard and it seems to be worth to try… unfortunately, 90% of the tags we use are coming from DiscoGS, not Musicbrainz.

Anyways, I doubt that local scanning (which seems to be a problem here), can be improved w/ MB, because on the Synology, I only want Plex to use local tags and embedded art.

Thanks for your help, I’ll wait for your response about the log.