I have a VPS server running only docker containers (nothing running aside of the docker daemon), these docker containers (all latest):
- gitlab - docs.gitlab.com/omnibus/docker/README.html
- a mailserver - GitHub - docker-mailserver/docker-mailserver: Production-ready fullstack but simple mail server (SMTP, IMAP, LDAP, Antispam, Antivirus, etc.) running inside a container.
- nextcloud - hub.docker.com/r/wonderfall/nextcloud/
- an OpenLDAP server - GitHub - osixia/docker-openldap: OpenLDAP container image 🐳🌴
- a webserver - hub.docker.com/_/httpd/ (+ custom Dockerfile)
- a nginx proxy for all the above - GitHub - nginx-proxy/nginx-proxy: Automated nginx proxy for Docker containers using docker-gen (+ custom Dockerfile)
With my VPS provider, I have access to some monitoring tools, in particular I can see the CPU load (%) and the memory usage (%) of my server.
I have a recurring issue where the server becomes “dead” - it does not respond anymore (ping / etc.). When this happens, the monitoring tools show a CPU usage of either ~25% (slightly higher) or 100% (it is a 4 CPUs VPS, so 25% would be one full CPU) BUT together with a RAM usage of 0%.
I spent hours trying to figure out the problem, trying to set logging programs in place, simply to notice that these stop logging just before the problem arises, so these are of no use.
I would like to know if anyone had experienced similar issues using docker and in particular one of the above mentioned containers? And if so, what could I do to fix these issues?