I’m trying to determine if I have something misconfigured or if I have found a bug. Running in production I have incus version 6.12 running on top of ubuntu 24.04. (This is running on a dell r640 with 384 GB of ram).
I have a docker container running inside an Incus container, the docker container is running Java and has a memory limit of 9GB, this was set in the docker run command. The Incus container running has a memory limit of 32GB.
Something happened that caused the Java application to lose it’s mind ( not only on this host but others as well) and for some reason docker did not kill the container when it went over 9GB of memory. As a matter of fact it didn’t stop until it got to the full amount of memory allocated to the incus container (32GB). On the other hosts where docker was NOT running in an incus container docker killed the java container when it hit 9GB.
