Arch Linux

DerJudge

Member

From: NRW, Germany

Registered: 2013-03-26

Posts: 24

Website

Cannot use Docker

Hi @all,

I am using Docker 17.06.0-ce from community/docker package:

# docker version
Client:
 Version:      17.06.0-ce
 API version:  1.30
 Go version:   go1.8.3
 Git commit:   3dfb8343
 Built:        Sat Jul  8 20:56:41 2017
 OS/Arch:      linux/amd64

Server:
 Version:      17.06.0-ce
 API version:  1.30 (minimum version 1.12)
 Go version:   go1.8.3
 Git commit:   02c1d87617
 Built:        Sat Jul  8 22:56:56 2017
 OS/Arch:      linux/amd64
 Experimental: false
#

When I try to start a container, I get this:

-- Unit postgres-9.3.service has begun starting up.
Jul 27 10:58:33 tango003.zen-net.de systemd-timesyncd[671]: Network configuration changed, trying to establish connection.
Jul 27 10:58:33 tango003.zen-net.de kernel: docker0: port 1(veth6be5671) entered blocking state
Jul 27 10:58:33 tango003.zen-net.de kernel: docker0: port 1(veth6be5671) entered disabled state
Jul 27 10:58:33 tango003.zen-net.de kernel: device veth6be5671 entered promiscuous mode
Jul 27 10:58:33 tango003.zen-net.de kernel: IPv6: ADDRCONF(NETDEV_UP): veth6be5671: link is not ready
Jul 27 10:58:34 tango003.zen-net.de systemd-udevd[16194]: link_config: autonegotiation is unset or enabled, the speed and duplex are not writable.
Jul 27 10:58:34 tango003.zen-net.de systemd-udevd[16194]: Could not generate persistent MAC address for veth17322f3: No such file or directory
Jul 27 10:58:34 tango003.zen-net.de systemd-udevd[16195]: link_config: autonegotiation is unset or enabled, the speed and duplex are not writable.
Jul 27 10:58:34 tango003.zen-net.de systemd-udevd[16195]: Could not generate persistent MAC address for veth6be5671: No such file or directory
Jul 27 10:58:34 tango003.zen-net.de kernel: eth0: renamed from veth17322f3
Jul 27 10:58:34 tango003.zen-net.de systemd-networkd[672]: veth6be5671: Gained carrier
Jul 27 10:58:34 tango003.zen-net.de systemd-networkd[672]: docker0: Gained carrier
Jul 27 10:58:34 tango003.zen-net.de kernel: IPv6: ADDRCONF(NETDEV_CHANGE): veth6be5671: link becomes ready
Jul 27 10:58:34 tango003.zen-net.de kernel: docker0: port 1(veth6be5671) entered blocking state
Jul 27 10:58:34 tango003.zen-net.de kernel: docker0: port 1(veth6be5671) entered forwarding state
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/cpu,cpuacct/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/cpu,cpuacct/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/net_cls,net_prio/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/net_cls,net_prio/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/freezer/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/freezer/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/devices/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/devices/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/cpuset/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/cpuset/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/pids/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/pids/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/blkio/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/blkio/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16218 to /sys/fs/cgroup/systemd/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 Moving pid 16249 to /sys/fs/cgroup/systemd/system.slice/postgres-9.3.service/cgroup.procs
Jul 27 10:58:34 tango003.zen-net.de systemd-docker[16166]: 2017/07/27 10:58:34 open /sys/fs/cgroup/system.slice/docker.service/cgroup.procs: no such file or directory
Jul 27 10:58:34 tango003.zen-net.de systemd[1]: postgres-9.3.service: Main process exited, code=exited, status=1/FAILURE
Jul 27 10:58:34 tango003.zen-net.de systemd[1]: Failed to trim compat systemd cgroup /system.slice/postgres-9.3.service: Device or resource busy
Jul 27 10:58:34 tango003.zen-net.de systemd[1]: Failed to start PostgreSQL server v9.3 (Atlassian supported).
-- Subject: Unit postgres-9.3.service has failed
-- Defined-By: systemd
-- Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit postgres-9.3.service has failed.
-- 
-- The result is failed.
Jul 27 10:58:34 tango003.zen-net.de systemd[1]: postgres-9.3.service: Unit entered failed state.
Jul 27 10:58:34 tango003.zen-net.de systemd[1]: postgres-9.3.service: Failed with result 'protocol'.
Jul 27 10:58:35 tango003.zen-net.de systemd-networkd[672]: veth6be5671: Gained IPv6LL

I have searched for this "open /sys/fs/cgroup/system.slice/docker.service/cgroup.procs: no such file or directory" - Error, and found that in most cases, it seems to be because of a too recent version of systemd, which docker is not supporting yet and that adding "--exec-opt native.cgroupdriver=cgroupfs" to dockerd's startup line should help.
So I did override this with this content in /etc/systemd/system/docker.service.d/override.conf :

[Service]
ExecStart=
ExecStart=/usr/bin/dockerd --exec-opt native.cgroupdriver=cgroupfs -H fd://

After that I reloaded systemd's inventory with "systemctl daemon-reload" and restarted docker.service.

I then tried to start my container using the following, but got the same results:

[Unit]
Description=PostgreSQL server v9.3 (Atlassian supported)
After=docker.service
Requires=docker.service

[Service]
ExecStart=/usr/bin/systemd-docker run --rm --name %n -e POSTGRES_PASSWORD=my_shiny_pass -p 127.0.0.1:5433:5432 -v /var/lib/postgres/9.3/data:/var/lib/postgresql/data postgres:9.3
Restart=always
RestartSec=20s
Type=notify
NotifyAccess=all
TimeoutStartSec=300
TimeoutStopSec=60

[Install]
WantedBy=multi-user.target

Does anybody have an idea on this?
Starting the container without systemd-docker (docker run --rm --name postgres-9.3.service -e POSTGRES_PASSWORD=my_shiny_pass -p 127.0.0.1:5433:5432 -v /var/lib/postgres/9.3/data:/var/lib/postgresql/data postgres:9.3) seems to work fine.