emby service goes offline very often

[Luke 42078]

Thanks for following up.

[LongMan 112]

No problem. Seems odd that another program could affect Emby is this way.

[LongMan 112]

Did the log provide any clue as to why this is happening? At the moment I have disabled auto scan in ersatztv. I manually trigger scans except for emby collections, but that is less than optimal because some of my channels are based on collections from Emby.

Any help here would be greatly appreciated.

Cheers,

LongMan

[Luke 42078]

Has the crashing stopped?

[LongMan 112]

No. It still occurs.

My workaround is to turn off auto scan on ersatz tv and only manually trigger the scan of the individual libraries. Scanning of collections has its own trigger which I avoid. So the collections from emby have not been updated for quite a while.

The channels based exclusively on collections from emby are not working.

BTW this started when I updated from 4.7 to 4.8

[Luke 42078]

Quote

My workaround is to turn off auto scan on ersatz tv and only manually trigger the scan of the individual libraries. Scanning of collections has its own trigger which I avoid. So the collections from emby have not been updated for quite a while.

So with this off does it still happen?

[LongMan 112]

Only if a trigger the collections scan. The log that i attached showed what happened when I triggered a collections scan

[sa2000 674]

14 hours ago, LongMan said:

The log that i attached showed what happened when I triggered a collections scan

Are you able to say what time you did that so can check the log entries after that ? How exactly was this triggered - for a specific collection? Would like to see if I can replicate the issue

I see some errors for user "ZotacUser" - i presume this is unrelated - it was through Firefox Windows on same machine as one that used Chrome Web for another Emby User. This is an extract of one of several similar logged errors

2024-06-17 09:03:05.464 Error Server: User ZotacUser does not have access to ManageServer feature.

Also there are some insufficient system resources errors being logged during backups - would be good to know if this was memory related and wonder if you can use tools such as SysInternals VMMap to monitor the Emby Server process memory usage  and see if the issue with collections scan is also memory related

These extracts were relating to the backups system resources related errors 

2024-06-17 00:11:15.969 Error Backup & Restore: Error getting file contents of \\Storage\ml vii 10t\EmbyBackups\embyserver-backup-full\metadata\library
	System.IO.IOException: System.IO.IOException: Insufficient system resources exist to complete the requested service. : '\\Storage\ml vii 10t\EmbyBackups\embyserver-backup-full\metadata\library\2e\2efa9f32a2c094b84826f243ee55be67'

2024-06-17 01:40:20.449 Error Backup & Restore: Error getting file contents of \\Storage\ml vii 10t\EmbyBackups\embyserver-backup-full\metadata\people
	System.IO.IOException: System.IO.IOException: Insufficient system resources exist to complete the requested service. : '\\Storage\ml vii 10t\EmbyBackups\embyserver-backup-full\metadata\people\D\Donovan Cook'

 

[LongMan 112]

First. Thanks for looking into this.
Are you looking at the this log? https://drive.google.com/file/d/1mtGOxAGoSYC87bCSzuMz46KDvG1cT9nV/view?usp=sharing

Search for PlayedMovies in the log. BTW ersatztv accesses emby via an API Key not a user.

Cheers,

LongMan

[sa2000 674]

Sorry I am not at my pc now. It is the log that I extracted those log lines I pasted from. At this moment I am not interested in what ersatztv does but in the manual scan you said you did with same outcome - 

[LongMan 112]

So that we are on the same page. It is a scan in ersatztv that syncs emby collections to it.

The problem is that another program can stop emby from working when it access the API.

 

[sa2000 674]

22 hours ago, LongMan said:

So that we are on the same page. It is a scan in ersatztv that syncs emby collections to it.

The problem is that another program can stop emby from working when it access the API.

Sorry I misunderstood. 

I can see these two requests and after the first arrived the server was not responding to requests. 

2024-06-17 08:32:13.188 Debug Server: http/1.1 GET http://172.24.145.7:14546/Items?parentId=449383&includeItemTypes=Movie,Series,Season,Episode&recursive=True&startIndex=0&limit=0. UserAgent: 
2024-06-17 08:32:13.189 Debug SqliteItemRepository: GetitemById BoxSet 449383 PlayedMovies

2024-06-17 08:33:53.209 Debug Server: http/1.1 GET http://172.24.145.7:14546/Items?parentId=398269&includeItemTypes=Movie,Series,Season,Episode&recursive=True&startIndex=0&limit=0. UserAgent: 
2024-06-17 08:34:51.729 Debug SqliteItemRepository: GetitemById BoxSet 398269 PlayedShows

I do not know how massive this request is and how many items get returned. Or whether it is just very slow and the database is locked out whilst the request is being processed.

If it is massive potentially returning many items, then it needs to be split as multiple requests making use of the StartIndex and Limit parameters

You could try the first request using curl in a command line window and see what you get back and how long it takes and also if memory usage is high.

The curl request would be like this

curl -o 449383_items.txt http://<ip-of-server>:<port>/Items?parentId=449383&includeItemTypes=Movie,Series,Season,Episode&recursive=True&startIndex=0&limit=0&api_key=xxxxxxxxxxx

put in the ip address of the server and your local port as well as a working api_key

curl will write the response to file 449383_items.txt in the local directory for the command line session you run it in

 

 

Edited June 28, 2024 by sa2000

[LongMan 112]

I will pass the info on.

Thank You Kind Sir.

Cheers,
LongMan

[Luke 42078]

On 6/28/2024 at 8:58 AM, LongMan said:

I will pass the info on.

Thank You Kind Sir.

Cheers,
LongMan

Please let us know how you get on. Thanks !

[LongMan 112]

@Luke
The problem has been resolved. The Developer was able to address it earlier today.

"ok, latest dev build removes the specific call identified in the post (startindex = 0, limit = 0)"

I triggered a manual scan and the server did not go offline and the collections have been synced to ersatz tv.

I am truly appreciative of your taking the time to look into this and to @sa2000 for the suggested solution.
 

Thank You.
LongMan