1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
# possible issues

1. docker container stops being able to read from the read only mount/volume?
2. ecs_agent stops being able to read from the dockercfg file?

   * https://github.com/remind101/amazon-ecs-agent/commits/master/Makefile




### bad controller.

ubuntu@ip-10-128-37-153:~$ docker run -it --volume /root/.dockercfg:/root/.dockercfg --entrypoint /bin/bash bb3e1fc70310

root@a83bd2dbb9a3:/go/src/github.com/remind101/empire# cat /root/.dockercfg

{"https://index.docker.io/v1/": {"email": "<removed>", "auth": "<removed>"}}



# after killing the docker image running on this host for 6 days, the ecs task scheduler
# rescheduled the task to this same bad host. The new container also had trouble accessing
# images on dockerhub which require authentication.

# killing this new process who was running for 15 minutes caused the ecs task scheduler
# to schedule a new ecs task on another controller host.
# the new task was able to download images from dockerhub which require authentication.



### good controller.

ubuntu@ip-10-128-25-57:~$ docker run -it --volume /root/.dockercfg:/root/.dockercfg --entrypoint /bin/bash bb3e1fc70310

root@c7a634b8c30d:/go/src/github.com/remind101/empire# cat /root/.dockercfg

{"https://index.docker.io/v1/": {"email": "<removed>", "auth": "<removed>"}}