Transferring Docker run script to compose file - postgres doesn't run? - postgresql

Question

I was previously running the Docker container for my web application below:
echo Starting postgres container...
docker run -d -p 5432:5432 --name db my_db
echo Postgres container started
echo Starting memcached container...
docker run -d -p 11211:11211 --name mem memcached
echo Memcached container started
echo Starting web container...
docker run -d -p 8000:8080 --name web --link db:postgres --link mem:memcache my_web_app
echo Web container started
echo Waiting for tomcat to initialize...
sleep 5
echo curling logon.jsf
curl http://111.111.11.111:8000/app/logon.jsf #Real URL redacted
echo Application started
Which became the below docker-compose script:
version: '2'
services:
web:
image: tomcat:7.0.69-jre8
build: ./ci
ports:
- "8000:8080"
depends_on:
- db
- mem
db:
image: postgres:9.4.7
build: ./db
ports:
- "5432:5432"
environment:
POSTGRES_USER: admin
POSTGRES_PASSWORD: admin
mem:
image: memcached
ports:
- "11211:11211"
When I launch the container, it seems like postgres is no longer running (Connection refused, netstat -al | grep postgres doesn't show it), and thus, the app doesn't work. I reviewed the change I have made, and realized that the docker-compose.yml has no analog for the tomcat sleep and curl command I had given in the run script - should it? I couldn't find any information online regarding a proper analog. Is something else potentially afoot here? Please advise if any further details are required.
EDIT: docker ps yields:
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
a942be02fc22 tomcat:7.0.69-jre8 "catalina.sh run" 5 minutes ago Up About a minute 0.0.0.0:8000->8080/tcp memcachedocker_web_1
fd6074cf8ec9 postgres:9.4.7 "/docker-entrypoint.s" 5 minutes ago Up About a minute 0.0.0.0:5432->5432/tcp memcachedocker_db_1
ce18771fee28 memcached "/entrypoint.sh memca" 5 minutes ago Up About a minute 0.0.0.0:11211->11211/tcp memcachedocker_mem_1
and the logs for the container:
Success. You can now start the database server using:
postgres -D /var/lib/postgresql/data
or
pg_ctl -D /var/lib/postgresql/data -l logfile start
waiting for server to start....LOG: database system was shut down at 2016-06-01 13:09:23 UTC
LOG: MultiXact member wraparound protections are now enabled
LOG: database system is ready to accept connections
LOG: autovacuum launcher started
done
server started
CREATE DATABASE
CREATE ROLE
/docker-entrypoint.sh: running /docker-entrypoint-initdb.d/init.sql
CREATE DATABASE
LOG: received fast shutdown request
LOG: aborting any active transactions
LOG: autovacuum launcher shutting down
LOG: shutting down
waiting for server to shut down....LOG: database system is shut down
done
server stopped
PostgreSQL init process complete; ready for start up.
LOG: database system was shut down at 2016-06-01 13:09:25 UTC
LOG: MultiXact member wraparound protections are now enabled
LOG: database system is ready to accept connections
LOG: autovacuum launcher started
and thus inside the container, :
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 127.0.0.11:44095 0.0.0.0:* LISTEN
tcp 0 0 172.18.0.4:38868 132.246.2.23:80 TIME_WAIT
tcp 0 0 172.18.0.4:36424 5.153.231.35:80 TIME_WAIT
tcp6 0 0 127.0.0.1:8005 :::* LISTEN
tcp6 0 0 :::8009 :::* LISTEN
tcp6 0 0 :::8080 :::* LISTEN
tcp6 0 0 172.18.0.4:47522 172.18.0.2:11211 ESTABLISHED
tcp6 0 0 172.18.0.4:47518 172.18.0.2:11211 ESTABLISHED
tcp6 0 0 172.18.0.4:47524 172.18.0.2:11211 ESTABLISHED
tcp6 0 0 172.18.0.4:47520 172.18.0.2:11211 ESTABLISHED
tcp6 0 0 172.18.0.4:47516 172.18.0.2:11211 ESTABLISHED
tcp6 0 0 172.18.0.4:47514 172.18.0.2:11211 ESTABLISHED
udp 0 0 172.18.0.4:48360 10.0.2.3:53 ESTABLISHED
udp 0 0 127.0.0.11:40276 0.0.0.0:*
We can see that the memcached container is up and running, however postgres is absent. The only thing running on the localhost interface is bound to 8005, and it isn't postgres. I should add finally that I am able to telnet memcached on its port from the web container and confirm its operation, however I can't do the same with postgres, and I'm not sure why.

Answer 1

There are a number of things that can probably be removed from your
docker-compose.yml file, assuming that the database and memcached
services are dedicated to your web application (that is, they don't
need to be accessed from anywhere else).
It doesn't make sense to specify both image and build in a service
definition, so I've made a few assumptions about what you actually
want.
I would start with something like this:
version: '2'
services:
web:
build: ./ci
ports:
- '8000:8080'
depends_on:
- db
- mem
db:
image: postgres
environment:
POSTGRES_USER: admin
POSTGRES_PASSWORD: admin
mem:
image: memcached
You don't need ports directives for either the db or mem
services, because you don't need to expose these services to your host
(or elsewhere). A docker container can access services on another
container on the same network without any special directives.
Within the web container, you can access either the db or mem
container by name, because this is how newer versions of Docker
operate when using a non-default network: Docker creates an internal
DNS service and registers the container names there. So, for example,
from inside the web container I can use the psql command line
postgresql client to connect to the database server like this:
$ docker exec -it myservice_web_1 sh
web_1# psql -h db -U admin
Password for user admin:
psql (9.4.6, server 9.5.1)
WARNING: psql major version 9.4, server major version 9.5.
Some psql features might not work.
admin=#
You would configure your web application to access the database server
at host db, port 5432.
With this environment up and running, you could connect to your web service from your host at http://localhost:8000, or from either your host or another host on the network at http://<your_hostname_or_ip_here>:8000.