Port forwarding with Elastic docker image - elasticsearch

Question

I'm trying to test docker out with this docker image. Things should be straight forward. But it isn't.
I ran this command to start the container:
sudo docker run -d -p 9200:9200 -p 9300:9300 elasticsearch -Des.node.name="ElasticTestNode"
Then I tried to run this command in my host machine:
# curl -XPUT "http://localhost:9200/movies/movie/3" -d'
{
"title": "To Kill a Mockingbird",
"director": "Robert Mulligan",
"year": 1962,
"genres": ["Crime", "Drama", "Mystery"]
}'
I was expecting to see some kind of successful message. Instead, the command simply stuck. No output and not stopping at all. I have to Ctrl-X to quit.
Ran out of idea, I started a bash shell inside the docker and tested:
sudo sudo docker exec -i -t some-docker-id /bin/bash
root#somehash:/usr/share/elasticsearch# curl -XPUT "http://localhost:9200/movies/movie/3" -d'
{
"title": "To Kill a Mockingbird",
"director": "Robert Mulligan",
"year": 1962,
"genres": ["Crime", "Drama", "Mystery"]
}'
{"_index":"movies","_type":"movie","_id":"3","_version":1,"_shards":{"total":2,"successful":1,"failed":0},"created":true}root#somehash:/usr/share/elasticsearch#
And it was a success. What have I done wrong?
Updates: Tried another command on my host machine:
$ curl -XPUT -v "http://localhost:9200/movies/movie/3" -d'
{
"title": "To Kill a Mockingbird",
"director": "Robert Mulligan",
"year": 1962,
"genres": ["Crime", "Drama", "Mystery"]
}'
* Trying 127.0.0.1...
* Connected to localhost (127.0.0.1) port 9200 (#0)
> PUT /movies/movie/3 HTTP/1.1
> Host: localhost:9200
> User-Agent: curl/7.43.0
> Accept: */*
> Content-Length: 139
> Content-Type: application/x-www-form-urlencoded
>
* upload completely sent off: 139 out of 139 bytes
Stuck here...
# sudo docker logs docker-id
[2016-09-28 11:52:16,630][INFO ][node ] [ElasticTestNode] version[2.4.0], pid[1], build[ce9f0c7/2016-08-29T09:14:17Z]
[2016-09-28 11:52:16,631][INFO ][node ] [ElasticTestNode] initializing ...
[2016-09-28 11:52:17,202][INFO ][plugins ] [ElasticTestNode] modules [reindex, lang-expression, lang-groovy], plugins [], sites []
[2016-09-28 11:52:17,219][INFO ][env ] [ElasticTestNode] using [1] data paths, mounts [[/usr/share/elasticsearch/data (/dev/sda8)]], net usable_space [5.4gb], net total_space [19.5gb], spins? [possibly], types [ext4]
[2016-09-28 11:52:17,219][INFO ][env ] [ElasticTestNode] heap size [990.7mb], compressed ordinary object pointers [true]
[2016-09-28 11:52:18,816][INFO ][node ] [ElasticTestNode] initialized
[2016-09-28 11:52:18,816][INFO ][node ] [ElasticTestNode] starting ...
[2016-09-28 11:52:18,877][INFO ][transport ] [ElasticTestNode] publish_address {172.17.0.22:9300}, bound_addresses {[::]:9300}
[2016-09-28 11:52:18,881][INFO ][discovery ] [ElasticTestNode] elasticsearch/LCo5k0dARimsWFXjN1Yu0A
[2016-09-28 11:52:21,915][INFO ][cluster.service ] [ElasticTestNode] new_master {ElasticTestNode}{LCo5k0dARimsWFXjN1Yu0A}{172.17.0.22}{172.17.0.22:9300}, reason: zen-disco-join(elected_as_master, [0] joins received)
[2016-09-28 11:52:21,924][INFO ][http ] [ElasticTestNode] publish_address {172.17.0.22:9200}, bound_addresses {[::]:9200}
[2016-09-28 11:52:21,925][INFO ][node ] [ElasticTestNode] started
[2016-09-28 11:52:21,960][INFO ][gateway ] [ElasticTestNode] recovered [0] indices into cluster_state

Answer 1

It seems that the port mapping of docker sometimes fail. I have experienced this issue multiple times. The same test script works on a boot while doesn't on another.
One thing consistence is that if things went bad on one boot, it stays fail every time I restart the docker image. It stays fail even after I ditched the container and start a new one with the image. It seems to be an issue of the docker daemon.
The way for me to solve this issue is to stop all container and restart the docker daemon:
sudo docker stop $(docker ps -a -q)
sudo systemctl restart docker
sudo docker start $(docker ps -a -q)
It works for me. Hope someone would find it helpful.