harsh river
Hello, I have a question about the speed of PMM. This is very different for me. Sometimes it takes a few seconds to parse the IDs of TMDb or Trakt, other times much longer. Every 10 to 30 or more seconds it goes one ID further.
The mass update also usually works faster since MultiBatch. Today, however, PMM again takes what feels like an eternity. A movie continues every 30 seconds ...
Is there any way to speed this up? What could be the reason for this?
tribal cloudBOT
Welcome @harsh river!
It looks like you have not yet completed #938455615741775902, this will allow us to help you quicker.
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.
tribal cloudBOT
You can press the "Close Post" button above or type `/close` at any time to clos...
harsh river
PMM sometimes extremely slow (Parsing ID, mass_update with MultiBatch)
sonic gust
What else is your server doing?
Are you running PIC, TCM, another instance of PMM, video thumbnails creation, credit detection, or intro detections?
What's your hardware doing? Local or remote?
Docker?
Ultimately we'd be guessing... with the amount of information you've given us.
harsh river
What is PIC or TCM?
I do not run any other PMM instance. Credit detection and Intro detections are used. However, they are only executed when new content is added.
Hardware is an Intel NUC with an i5 cpu. Does nothing except run Plex Server. The computer is not used for other services.
The PMM script is executed locally directly on the computer where Plex is running.
Docker is not used. This is a Windows 11 machine.
kind slate
Can you provide a log of this happening?
harsh river
harsh river
Can you provide a log of this happening?
For sure 🙂
tribal cloudBOT
📝 Great! Let's start to review and make recommendations, dead_search... 📝
📝 If you want to review this again, dead_search:
: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> 📝
kind slate
Recent changes in Plex have made API operations much slower in recent releases, including 1.32.8:
Connected to server Antares version 1.32.8.7639-fb6452ebf
NIghtly 30 only contains batch collection adds, so all these operations are being done individually, which is probably the issue.
harsh river
I have already imported the multiBatch_operations manually.
harsh river
I'm sorry. I forgot to mention that. I only wrote in the initial post that I use MultiBatch - but not which one. :/
But parsing the IDs from trakt, mdblist or tmdb also takes an extremely long time in some cases. Each ID feels like an eternity.
slender torrent
I see what you mean in your logs, going from one movie to the next is taking a long time. Unfortunately this is before any of the batchMultiEdit stuff even gets invoked. If you run the operations again with the -lr flag it should show us exactly which plex requests are taking time.
Like Mr Carrot said though, some plex operations are just slow on recent versions. Sometimes I see issues if Plex is doing other things at the same time like scanning media or detecting intros, could be that?
harsh river
I see what you mean in your logs, going from one movie to the next is taking a l...
Ok, I will do that! Thank you!
harsh river
I see what you mean in your logs, going from one movie to the next is taking a l...
Automatic scanning is deactivated. Intro and credits are created directly when adding media. But of course I can't see what else is being done by Plex - that could be the reason. Yes 😦
slender torrent
do you use something like autoscan though to trigger scans when media is added?
harsh river
No, this is completely disabled in Plex. I have to initiate every scan manually.
Also, parsing has nothing to do with Plex, right? This is about loading the information from tmdb, mdblist, trakt etc.?
slender torrent
it might be comparing things from tmdb (for example) with those in plex, which might mean requests are going to plex for that information. The -lr flag will tell us
harsh river
All right! Can I set PMM to always use the -lr flag?
harsh river
PMM_LOG_REQUESTS=true in yml file 👍
kind slate
"yml file"? If you mean a docker-compose yml, then sure.
slender torrent
no, either in the terminal or in your docker env variables
or cron if you're scheduled that way
harsh river
All right. So I have to adapt my batch files.
verbal bronze
Here is the first part of the log file. It takes much longer again. The first time multiBatch_operations was run, it was extremely fast - one run took 30 minutes. Now it will probably take 6 hours again...
tribal cloudBOT
📝 If you want to review this again, deadsearch:
: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> 📝
sonic gust
Why the "alt" accounts?
harsh river
I beg your pardon. Apparently I still have my old Discord account active on my other computer. I will change it immediately.
slender torrent
Hey @verbal bronze - looking at the logs it seems that it's these calls that are taking their time:
http://192.168.188.206:32400 "GET /library/metadata/67972?includePreferences=1 HTTP/1.1" 200 3246
That one for example looks like it took around 7 seconds. What this is doing is retrieving the current information about a movie so that it can then make a comparison with TMDB/IMDb etc and decide what needs to be updated. This process hasn't changed with the introduction of batchMultiEdits, and the slowness is due to the response from Plex.
If it was quicker before, perhaps it could be worth restarting your Plex instance, or perhaps seeing if ChuckPa's DBRepair script could offer any performance benefits
Or indeed if you want to try the latest master build for comparison...we could see if it's something in my recent changes. It doesn't look like it from here though.
sonic gust
I beg your pardon. Apparently I still have my old Discord account active on my o...
This was a little weird...that's all. Was not sure if there were two different people or not.
It's also probably an intermittent thing... meaning that it might go away as possibly Plex was busy doing something when pmm was asking it for information. Moving back to master might make it better or worse or exactly the same.
harsh river
Hey <@456226577798135808> - looking at the logs it seems that it's these calls t...
Thank you for your answer! I forgot to say thank you - sorry! 😬 I also think it's because of Plex ... Today two passes slipped through in a few minutes.
slender torrent
Cool!
sonic gust
Thank you for your answer! I forgot to say thank you - sorry! 😬 I also think it...
Anything else needed here?
harsh river
Thank you, nothing else needed here.