Beanstalkd in docker crashes due to lack of memory. Why?

Question

linux

Hello! I am using beanstalkd queue server. More than 100 thousand tasks are processed in the queue in an hour, everything is fine, but after 3-4 hours of work, the docker container with beanstalkd flies off due to lack of RAM...
Oct 15 13:52:40 vps kernel: Out of memory: Kill process 3642 (beanstalkd) score 671 or sacrifice child
I monitor the container with the "htop" command, the load in it is 0.1-4% of the RAM, and so on throughout its work. What could be causing the process to terminate?
I start beanstalkd with the command:
docker run -d -p 11300:11300 schickling/beanstalkd -z 1512000
Workers work through supervisor

Answer 1

I monitor the container with the "htop" command, the load in it is 0.1-4% of the RAM, and so on throughout its work

Something I can't believe. Do you have a monitoring cronjob or do you run it manually from time to time?
I would do a cron like
top -b -o +%MEM | head -n 10 >> /tmp/memory.out
I'm sure we'll see a memory leak.
As for the idea of ​​disabling OOKiller - I don't know exactly what will happen, but I'm sure it won't improve the situation.