Open danielbjornadal opened 6 years ago
@danielbjornadal. was able to run this test with docker CE 18.03.0 and unable to repro the issue. But from your logs looks like the plugin is able to report the status of the volume accurately (attached/detached). Suggest opening an issue on Docker as this doesn't seem to be a plugin issue.
@govint. We upgraded to 18.03.0-ce, and the issue still persist. We have tried on both debian and ubuntu-xenial.
# docker version
Client:
Version: 18.03.0-ce
API version: 1.37
Go version: go1.9.4
Git commit: 0520e24
Built: Wed Mar 21 23:10:06 2018
OS/Arch: linux/amd64
Experimental: false
Orchestrator: swarm
Server:
Engine:
Version: 18.03.0-ce
API version: 1.37 (minimum version 1.12)
Go version: go1.9.4
Git commit: 0520e24
Built: Wed Mar 21 23:08:35 2018
OS/Arch: linux/amd64
Experimental: false
# docker exec -it voltest1 bash
OCI runtime exec failed: exec failed: container_linux.go:348: starting container process caused "exec: \"bash\": executable file not found in $PATH": unknown
Is there any debugging we could try on the ESXi host?
@danielbjornadal, is this the problem,
docker ps -a CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES b8c51d612bd7 busybox "sh" 5 seconds ago Up 4 seconds voltest1 4f55637b76c2 busybox "sh" 18 hours ago Up 18 hours voltest
docker exec -it b8c51d612bd7 bash OCI runtime exec failed: exec failed: container_linux.go:348: starting container process caused "exec: \"bash\": executable file not found in $PATH": unknown
docker exec -it b8c51d612bd7 /bin/bash OCI runtime exec failed: exec failed: container_linux.go:348: starting container process caused "exec: \"/bin/bash\": stat /bin/bash: no such file or directory": unknown root@sc-rdops-vm02-dhcp-52-237:~# docker exec -it b8c51d612bd7 /bin/sh
/ # ps PID USER TIME COMMAND 1 root 0:00 sh 13 root 0:00 /bin/sh 18 root 0:00 ps / #
bash doesn't exist in the busybox image
@govint My bad. I was so focused on the problem, I did not check that busybox had bash.
The problem started with using this plugin through Rancher. If you install the plugin on the Rancher Agent host, and try to start a container using the vpshere volume plugin (using ubuntu:xenial this time :), the container is stuck in a starting state.
If I then try to to bash into the container from the host, it works. Don't know if this is a plugin or Rancher problem, but it looks like that Rancher can't see the container status.
# docker exec -it voltest3 bash
root@aab205625af5:
Rancher Server: v1.6.17-rc3 Rancher Agent: v1.2.10
@danielbjornadal ok, without Rancher this problem doesn;t seem to be an issue. Could you check with Rancher on this issue first. If on Rancher the plugin shows the proper status attached/detached at least we know the plugin is mounting/unmounting the volume whenever it's requested to do so. I don't believe this is a plugin issue.
Rancher issue: https://github.com/rancher/rancher/issues/12863 Plugin attach and detach the volume as it should.
Intro
After creating a vsphere volume, and mounting it to a container for the second time, the container is stuck in starting state.
Versions
ESXi VIB Driver: 0.21.2
Container stuck in starting state
# docker run -d -t --name voltest --volume-driver vsphere -v MyVolume2@t-ESXi.Data01:/mnt busybox
Container stopped