PMM Crashing Plex | Kometa | Page 1

rocky parrotBOT Jan 4, 2024, 10:00 PM

#

Welcome @potent sonnet!

Someone from <@&938443185347244033> will assist when they're available.

Including the meta.log from the beginning is a huge help, type !logs for more information.

#

You can press the "Close Post" button above or type /close at any time to close this post.

rocky parrotBOT Jan 4, 2024, 10:00 PM

#

rocky parrot You can press the "Close Post" button above or type `/close` at any time to clos...

#

📝 If you want to review this again, legion7:
:one: Right-click (or long press with phone) on the message that contains the log
:two: Select: Copy Message Link
:three: Use the command: /logscan <message_link> or !logscan <message_link> and paste the value copied from the previous step where you see <message_link> 📝

#

**Rec 15** - 💬🪟🐧 **WSL MEMORY RECOMMENDATION**

💬🪟🐧 WSL MEMORY RECOMMENDATION
According to Microsoft’s documentation, the amount of system memory (RAM) that gets allocated to WSL is limited to either 50% of your total memory or 8GB, whichever happens to be smaller.

It is possible to override the maximum RAM allocation, we suggest googling 'WSL memory limit' to learn more otherwise the following may work for you:To override the maximum RAM allocation when running Windows Subsystem for Linux (WSL), you need to modify the configuration settings. Here are the steps to do this:

Open a PowerShell window as an administrator.
Run the command: wsl --set-default-version 2 to set WSL version to 2 (WSL 2).
Run the command: wsl --set-memory <your_memory_limit> to set the maximum memory limit for WSL (replace <your_memory_limit> with the desired memory limit, e.g., 4GB).
Restart WSL by running the command: wsl --shutdown.

It is important to note that modifying these settings may require a reboot of your system.

runic lance Jan 4, 2024, 10:26 PM

#

if you don't use anime, you can take that out anidb

vivid shadow Jan 4, 2024, 10:34 PM

#

There are lots of these:

[ERROR]    | Unknown Error: (404) not_found; (redacted)/library/sections/1/all?id=118613&summary.locked=1&summary.value=Movies%20produced%20by...

Happening when it tries to update the summary of collections; often this is an issue with differing case in collection names [Plex has "My cool collection", PMM is trying to find "My Cool Collection"].

[ERROR]    | Collection Error: trakt_list requires Trakt to be configured                                       |

Trakt isn't fully configured.

[ERROR]    |     Overlay Error: This item's poster already has an Overlay. There is no PMM setting to change; manual attention required. |

There are a half-dozen cases where PMM skips applying overlays because the item's poster is already an overlaid poster that came from PMM but the item doesn't have the Overlay label on it.

[ERROR]    |   Overlay Error: No poster found                                                                   |

4 things don't have posters to apply overlays on.

[ERROR]    | Unknown Error: HTTPConnectionPool(redacted): Read timed out. (read timeout=60)                     |

While working on "Golden Globe 2023 Collection", plex starts timing out; recovers a couple minutes later on 2019.

Then starts timing out again while applying TV overlays, which eventually kills that process.

Then playlists all fail with timeouts, which is presumably when Plex had gone unresponsive.

#

You mean like the library name doesn't match?

No, the collection name. PMM is asking Plex to update the summary on a collection, and Plex is replying "I don't have one of those".

is there a way to wipe

Yes. There's an overlay reset script, and if you run PMM with the -dc runtime flag it will delete all collections when it starts processing a library.

#

Not sure what "docker command" you mean, but if you are using docker run -it -v "whatever:/config:rw" meisnate12/plex-meta-manager --run, yes, just add -dc onto the end there.

https://metamanager.wiki/en/nightly/pmm/environmental/#__tabbed_20_2

rocky parrotBOT Jan 5, 2024, 1:45 PM

#

📝 If you want to review this again, legion7:
:one: Right-click (or long press with phone) on the message that contains the log
:two: Select: Copy Message Link
:three: Use the command: /logscan <message_link> or !logscan <message_link> and paste the value copied from the previous step where you see <message_link> 📝

#

**PMM Info**

PMM Header

Version: 1.19.1 (Docker)
PlexAPI library version: 4.15.2
Platform: Linux-5.15.133.1-microsoft-standard-WSL2-x86_64-with-glibc2.28
Memory: 31 GB

Run Command: plex_meta_manager.py

PMM Footer

Finished 05:00 Run
Version: 1.19.1
Finished: 05:06:38 2024-01-05 Run Time: 0:06:01

Available PMM Versions

as of 2024-01-05 14:45:14

PMM Version (Master branch)

1.19.1

PMM Version (Develop branch)

1.19.1-develop78

PMM Version (Nightly branch)

1.19.1-nightly82

sage oyster Jan 5, 2024, 1:46 PM

#

Move to develop please and retry

digital ingotBOT Jan 5, 2024, 1:46 PM

#

Moving between branches of PMM

NOTE: This is addressing changing from, for instance, latest to nightly.

The method depends on how you installed it.

**I'M USING DOCKER:**

docker pull meisnate12/plex-meta-manager:TAG_HERE

replace TAG_HERE with latest, develop, or nightly; whichever you want to switch to.

Then recreate your container via whatever means you used to create it [docker run, docker-compose, etc.], setting the image as meisnate12/plex-meta-manager:TAG_HERE

**I'M USING DOCKER ON UNRAID**

:one: Click on DOCKER tab and then on Plex-Meta-Manager
:two: Edit Repository: setting to switch branches:

meisnate12/plex-meta-manager:TAG_HERE

replace TAG_HERE with latest, develop, or nightly; whichever you want to switch to.

**I'M USING DOCKER ON A SYSTEM WITHOUT COMMAND PROMPT (SUCH AS NAS):**

The same concepts as above will apply:
:one: pull new image
:two: recreate container using that image instead of the current one.

Whatever the system, there will be a way to accomplish these two things.

**I'M RUNNING LOCALLY VIA GIT:**

cd DIRECTORY_WHERE_YOU_PUT_PMM
git checkout BRANCH_NAME
git stash
git stash clear
git pull

For BRANCH_NAME, substitute master, develop, or nightly

Those stash commands are going to discard any changes you may have made to PMM's own files. If that's a concern, you are probably familiar enough with git to know what you need to do instead.

sage oyster Jan 5, 2024, 1:56 PM

#

Assuming it does not die after 6 minutes like the last run you shared.

sage oyster Jan 5, 2024, 6:32 PM

#

good sign i guess

rocky parrotBOT Jan 5, 2024, 11:03 PM

#

📝 If you want to review this again, legion7:
:one: Right-click (or long press with phone) on the message that contains the log
:two: Select: Copy Message Link
:three: Use the command: /logscan <message_link> or !logscan <message_link> and paste the value copied from the previous step where you see <message_link> 📝

#

**Rec 06** - ❌⏱️ **TIMEOUT ERROR**

❌⏱️ TIMEOUT ERROR
There were timeout issues while trying to connect to different services.
Ensure that your network configuration allows PMM to make internet calls.
Typically this is your Plex server timing out when PMM tries to connect to it. There's nothing PMM can do about this directly. Currently your timeout for plex is set to: 360 seconds. You can try increasing the connection timeout in config.yml:

  url: http://bing.bang.boing
  token: REDACTED
  timeout: 360   <<< right here```
But that's not a guarantee.

Effectively what's happening here is that you're ringing the doorbell and no one's answering. You can't do anything about that aside from waiting longer. You can't ring the doorbell differently.

This seems to happen most often in an Appbox context, so perhaps contact your appbox provider to discuss it.

In the PMM discord thread, type `!timeout` for more information.
For more information on network configuration, [https://www.metamanager.wiki/en/latest/home/installation.html]
6 line(s) with timeout errors. Line number(s): 85953, 86002, 86049, 86090, 86137, 86178

#

📎 parsed_meta.log_config_LegioN7.yml

runic lance Jan 5, 2024, 11:05 PM

#

incomplete logs

vivid shadow Jan 5, 2024, 11:07 PM

#

If it's exiting without an error like that, something external is killing the script. A likely culprit is out-of-memory

#

Just offering suggestions.

#

errors are pretty boring until Plex starts timing out and then the script exits.

[2024-01-05 13:49:18,159] [config.py:1123]            [ERROR]    | Invalid Root Folder: 'G:/Movies' Options: [[6:G:\Movies], [7:I:\Movies (2)]]                       |
[2024-01-05 13:49:18,210] [config.py:1155]            [ERROR]    | Invalid Root Folder: 'E:/TV Series' Options: [[5:E:\TV Series], [7:F:\Anime], [10:H:\TV Series (3)], [11:J:\Cartoons]] |

[2024-01-05 13:49:21,288] [config.py:1123]            [ERROR]    | Invalid Root Folder: 'G:/Movies' Options: [[6:G:\Movies], [7:I:\Movies (2)]]                       |
[2024-01-05 13:49:22,799] [config.py:1155]            [ERROR]    | Invalid Root Folder: 'E:/TV Series' Options: [[5:E:\TV Series], [7:F:\Anime], [10:H:\TV Series (3)], [11:J:\Cartoons]] |

[2024-01-05 21:47:03,313] [overlays.py:368]           [ERROR]    |     Overlay Error: HTTPConnectionPool(redacted): Read timed out. (read timeout=360)                |
[2024-01-05 22:23:53,964] [overlays.py:368]           [ERROR]    |     Overlay Error: HTTPConnectionPool(redacted): Read timed out. (read timeout=360)                |
[2024-01-05 23:00:44,470] [overlays.py:368]           [ERROR]    |     Overlay Error: HTTPConnectionPool(redacted): Read timed out. (read timeout=360)                |
                                                                 |   Overlays Attempted on 90 Day Fiancé: The Other Way S03E09: Overlay File (1) Dolby-Digital-Plus, Overlay File (2) Episode Info, Overlay File (4) 720P |
[2024-01-05 23:00:44,530] [overlays.py:180]           [INFO]     |                                                                                                    |
                                                                 | 90 Day Fiancé: The Other Way S03E10                                                                |

#

Well, if Plex is busy doing other stuff it's going to have less time available to respond to API requests, which may cause timeouts.

#

I don't think it's PMM that is causing Plex to crash here.

runic lance Jan 5, 2024, 11:14 PM

#

right, so hopefully you aren't trying to run pmm while maintenance or other scripts are running

vivid shadow Jan 5, 2024, 11:15 PM

#

All we can tell here is that things were going fine until Plex stopped responding and then something killed PMM.

runic lance Jan 5, 2024, 11:17 PM

#

probably on this same thread

#

like what are you wanting to have suggested?

#

so what are you proposing to add?

vivid shadow Jan 5, 2024, 11:22 PM

#

It's already falling down applying all the episode-level stuff and you want to give it more to do?

runic lance Jan 5, 2024, 11:22 PM

#

i was just going to say that, cause ratings is pretty intensive

vivid shadow Jan 5, 2024, 11:22 PM

#

you haven't asked about trimming it down; you've asked how to add more to it.

#

adding more stuff is unrelated to trimming it down, so I'm not sure how adding more helps you learn about removing things.

#

But, yes, if you want to add ratings overlays to movies the above will do that. It will also leave all the other overlays in place.

#

Well, except that the indentation is wrong.

runic lance Jan 5, 2024, 11:26 PM

#

you'd have to remove all other overlays from your config

vivid shadow Jan 5, 2024, 11:27 PM

#

Note the one that is not like the others

#

I think issue #1 should be "why is Plex not responding?"

#

then "who's killing PMM mid-run?"

#

Personally I'd get rid of all those episode-level overlays, but I'm a curmudgeon who doesn't think they provide value commensurate with the expense in applying them

#

I guess I'll skip this episode of Westworld because it's not the same audio codec as the previous one.

sage oyster Jan 5, 2024, 11:40 PM

#

If I recall, you were getting busy db stuff

#

200ms issues

#

So... this sounds like you are trying to make your Plex server do too much at the same time.

#

Meaning.... maintenance and PMM could be running at the same time and then that is "too much"

#

Refreshing metadata, scanning for files, preview thumbnails, credit and intro detection

#

All that will affect Plex db and performance

#

What time is Plex maintenance scheduled for?

#

Plex defaults at 2-5am

#

Starting PMM at 5am is fine

#

Are you using TCM (Titlecardmaker)

#

Do you have Plex auto scan?

#

I have seen that sometimes cause Plex timeouts.

#

Ok

#

Plex-auto scan is an app

#

Right?

#

Ok. Cool. It's a 3rd party adding to Plex

#

It can sometimes make things worse

#

Ok... then the other thing you may want to consider are the following performance tricks

digital ingotBOT Jan 5, 2024, 11:46 PM

#

PMM Performance Tips

Use PMM caching where possible: https://metamanager.wiki/en/latest/search.html?q=cache&check_keywords=yes&area=default

Run PMM after PLEX scheduled tasks so 5am is a good time for that since PLEX runs scheduled tasks between 3-5am

Consider Database Cache Size (MB) settings in PLEX under Settings, Library, Database Cache Size (MB). 2GB = 2048MB

Look into using the ChuckPA dbrepair tool to deal with plex db issues/fixes(He is an employee of Plex)

sage oyster Jan 5, 2024, 11:47 PM

#

Not sure if we already shared the above Dbrepair tool

#

Oh... you have my kitchen sink setup

#

lol

#

Until you figure out what is causing the crashes, you should not

#

Have you ever gotten through a successful run?

#

Which branch are you using?

#

Can you share the OOM of your Plex server?

#

Yes... absolutely

#

Batch edits will save you

#

OOM - Order of Magnitude

#

Switch to develop before your next run

#

How many media items in total is what I was looking for

digital ingotBOT Jan 5, 2024, 11:58 PM

#

Moving between branches of PMM

NOTE: This is addressing changing from, for instance, latest to nightly.

The method depends on how you installed it.

**I'M USING DOCKER:**

docker pull meisnate12/plex-meta-manager:TAG_HERE

replace TAG_HERE with latest, develop, or nightly; whichever you want to switch to.

Then recreate your container via whatever means you used to create it [docker run, docker-compose, etc.], setting the image as meisnate12/plex-meta-manager:TAG_HERE

**I'M USING DOCKER ON UNRAID**

:one: Click on DOCKER tab and then on Plex-Meta-Manager
:two: Edit Repository: setting to switch branches:

meisnate12/plex-meta-manager:TAG_HERE

replace TAG_HERE with latest, develop, or nightly; whichever you want to switch to.

**I'M USING DOCKER ON A SYSTEM WITHOUT COMMAND PROMPT (SUCH AS NAS):**

The same concepts as above will apply:
:one: pull new image
:two: recreate container using that image instead of the current one.

Whatever the system, there will be a way to accomplish these two things.

**I'M RUNNING LOCALLY VIA GIT:**

cd DIRECTORY_WHERE_YOU_PUT_PMM
git checkout BRANCH_NAME
git stash
git stash clear
git pull

For BRANCH_NAME, substitute master, develop, or nightly

Those stash commands are going to discard any changes you may have made to PMM's own files. If that's a concern, you are probably familiar enough with git to know what you need to do instead.

sage oyster Jan 6, 2024, 6:27 AM

#

So similar in number to what I have.

#

I have about 350 movies and about 4000 episodes

sage oyster Jan 6, 2024, 2:54 PM

#

In need of repair?

#

Use the ChuckPA dbrepair tool

#PMM Crashing Plex