It is as if the CMD has no effect.
In my ECS task definition, I have defined the command as seen in the below screenshot
The python process is supposed to be a 'blocking process' - it supposes to wait on data in a SQS queue.
However from the cloud watch log, it seems like the task gets spawned continuously
In effect it is like executing this
docker run -t simplequeue python /docverter/app/src/main.py
The container is started and then terminated right away.
I have defined the ENTRYPOINT in my dockerfile
FROM ubuntu:16.04
RUN apt-get update -y --fix-missing
RUN apt-get install -y git python python-pip cron ntp
ENV APP_HOME /docverter
RUN mkdir -p ${APP_HOME}
COPY . ${APP_HOME}
#
# Log configuration
RUN mkdir -p /root/.aws
RUN mkdir -p /var/awslogs/state
COPY ./credentials /root/.aws/credentials
COPY docker-entrypoint.sh /docker-entrypoint.sh
RUN apt-get install -y curl
RUN curl https://s3.amazonaws.com/aws-cloudwatch/downloads/latest/awslogs-agent-setup.py -O
RUN python awslogs-agent-setup.py --non-interactive -c ${APP_HOME}/aws-log.cfg --region ap-southeast-2
# Entry point
# To ensure some background services are started when the container is started
ENTRYPOINT /docker-entrypoint.sh
And the docker-entrypoint is as follow:
#!/bin/bash
set -eo pipefail
service awslogs start
service ntp start
I want to know how I can make the docker container to execute the Command in the task definition
EDIT as per Andrey's answer, I have modified the entrypoint script but it does not solve the issue. I added an extra echo "$@" for debug and it prints blank.
docker build -t broken .
docker run -it broken python # export to have python started
The test code can be cloned from here: git clone -b broken-entry-point-and-cmd https://github.com/kongakong/aws-ecs
