Hi everyone,
I’m experiencing a performance issue with the xibo-cms-db-1 container, which runs MySQL 8.0 as part of a Xibo CMS 4.3.0 setup using Docker Compose on Ubuntu 22.04.
The VPS has 4 CPU cores and 8 GB of RAM.
I’ve already increased the container’s mem_limit to 2 GB, but CPU usage remains extremely high — reaching 100% consistently, even with a current load of 230 active players.
I’ve checked:
MySQL logs (no critical errors, only warnings about --pid-file and deprecated --skip-slave-start)
Scheduled tasks in the CMS (no visible failures)
SHOW FULL PROCESSLIST (no locked or looping queries)
docker stats confirms sustained high CPU and memory usage in cms-db
I’d appreciate any help or guidance to diagnose and resolve this behavior.
Could those 200+ players be querying too much to DB with collection interval time ?
240/60=4 query from displays to DB per second if 1 minute interval, that does not sound too much ?
In CMS check display setting for collect interval, for test, I would raise collect interval time to 5-10 minutes and then check CPU usage.
I have several players querying on 1 min collect interval:
Does your DB memory usage 50% using 1 GB seems OK ?
Maybe setup standalone external DB and connect to external DB instead of dockerized one to compare CPU/MEM usage ?
After increasing the query interval, CPU usage dropped dramatically from 100% to just 4% within a few hours. This VPS belongs to a client, and until a few days ago, its average usage was around 10%. The issue began when he enabled report collection, which caused a system overload. Once we adjusted the players’ query timing to the server, CPU usage returned to normal.
The suggestion to use a database is an excellent one — I’ll discuss that possibility with my client. Thank you very much for your help!
