I have Jenkins
to build my project and run my unit test and integration test. Before all this, Jenkins
is launching several docker containers (with mongodb
, cassandra
etc inside) and then it starts. Sometimes, my tests fail because the they cannot reach the resources. After digging a bit I noticed that not all my containers start. So here is what I have:
docker-compose.yml
with the definition of each docker
container
start_docker.sh
script:
time docker-compose pull
time docker-compose --project-name $JOB_NAME up -d
time docker-compose --project-name $JOB_NAME ps
echo "Wait services are started"
docker-compose --project-name $JOB_NAME ps -q
container_names=`docker-compose --project-name $JOB_NAME ps -q`
container_nb="${#container_names}"
for container_name in $container_names; do
ports=`docker port $container_name | cut -d "/" -f1 `
service_ip=`docker inspect $container_name | grep "IPAddress" | grep 172 | sed "s/[^0-9]*\\([0-9\\.]\\+\\)[^0-9]*/\\1/"`
for port in $ports; do
while ! nc -z $service_ip $port; do
sleep 1 # wait for 1 second before check again
done
done
done
What I've noticed is that sometimes the ports
variable is empty for some containers and if I'm displaying them here is what I see:
[unit_test] target2sellcoredevelop_dse_1 /etc/dse/run.sh Up
[unit_test] target2sellcoredevelop_mongo_1 docker-entrypoint.sh mongo ... Up
[unit_test] target2sellcoredevelop_pentaho-pdi_1 /bin/sh -c /bin/bash -c "/ ... Up 0.0.0.0:9999->9999/tcp
[unit_test] target2sellcoredevelop_rabbitmq_1 docker-entrypoint.sh /init.sh Up
[unit_test] target2sellcoredevelop_redis_sentinel1_1 docker-entrypoint.sh sh -c ... Up
[unit_test] target2sellcoredevelop_rediscachereco_master_1 docker-entrypoint.sh redis ... Up
[unit_test] target2sellcoredevelop_rediscatalog_master_1 docker-entrypoint.sh redis ... Up
[unit_test] target2sellcoredevelop_redisuser_master_1 docker-entrypoint.sh redis ... Up
As you can see some of them already have the port number and some of them don't. I have 2 problems:
Why the port number is not available? Is because some docker zombie process? If so, how I can clean the up?
How can I properly wait for all them to really start?
EDIT: I think my problem is that sometimes some containers does not start. My question: how do I investigate that ?
You have a number of different questions, but in Docker Compose, you can wait for certain containers to come up before others using dadarek/wait-for-dependencies .
1). Add a new service to your docker-compose.yml
waitfordb:
image: dadarek/wait-for-dependencies
depends_on:
- mongodb
command: mongodb:27017
2). Add the following config to the service that requires MongoDB to be up. This will essentially wait for MongoDB to be fully up.
depends_on:
- waitfordb
3). Startup compose
docker-compose run --rm waitfordb
docker-compose up -d <SERVICE_1> <SERVICE_2>
This may or may not resolve your other issues, but might be a good place to start.
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.