docker PIDS keeps increasing at 7.0.8

[tamaszsellerp92]

I'm running a plain and simple 7.0.8 image with a modified config

storage:
dbPath: /data/db
directoryPerDB: true
engine: wiredTiger
wiredTiger:
engineConfig:
directoryForIndexes: true

and the PIDS number in docker stats keeps increasing rapidly: 14696 for a 3 day old container

CONTAINER ID   NAME             CPU %     MEM USAGE / LIMIT     MEM %     NET I/O          BLOCK I/O         PIDS
230acbbcf833   serene_mestorf   101.39%   7.859GiB / 23.47GiB   33.49%    50.7GB / 481GB   6.46GB / 17.4GB   14696

the ps within the docker container shows only 3 processes, their pids are constant

what is going on? only the mongodb container is raising the PIDS number.
Healthcheck is not enabled on this container.

[tianon]

That is definitely an extremely large number -- I don't think it's even cumulative. 😬

The difference between ps/top and that number is probably threads -- if you use something like docker top mongo -eLf you'll probably (hopefully) see a much longer list.

Is this instance exposed to the internet? I think every incoming connection generates a thread, and that could account for such a huge number (but would also be terrifying 😬).