Docker - the open-source application container engine. This chat is intended for contributors new to the Docker project or new to open source. For user help, please goto #docker on freenode. For advanced contributors please goto #docker-dev on freenode.
Hello all, I have quite an interesting issue with Docker Swarm, and I'm wondering if someone can help.
It seems as though my Docker Swarm service that I'm running is hitting my DNS server wayy too hard and causing timeouts/failures. Every once and a while, I get a
hostname resolving error (lookup <my-database-ip> on 127.0.0.11:53: read udp 127.0.0.1:35818->127.0.0.11:53: i/o timeout)
It seems to only happen on a single node, although it is intermittent and will sometimes cause other nodes problems. I'm not sure how to look further into this, and I'm wondering if anyone has had similar problems
Phase
> <@gitter_niuziyuanpy:matrix.org> [graphdriver] prior storage driver overlay2 failed: driver not supported
If I remember correctly, I got this after an upgrade of docker. Luckily, thx to the upgrade, they were no container needed on the machine. Thus, I moved the content of /var/lib/docker to somewhere else. A reboot of the docker daemon later, the daemon started correctly. Be careful, it had lost all its images.
Haley
Only way to cope with that, is:
Haley
1. Before the upgrade of the OS/Docker version, save all layers/images you need as tar files using the save command.
Haley
2. upgrade the OS, but remove docker
Haley
3. delete /var/lib/docker
Haley
4. install docker again (new version). make sure it runs.