Example Process Specs 

Contents

Example Process Specs

Web Server 

The file examples/nginx.json is the process spec for a simple “hello world” nginx webserver.

Launching the Process 

# From the root of the repo
bin/hpctl start examples/nginx.json

Making Sure It Worked 

$s in the code block below denote commands; the rest is example output.

$ bin/hpctl list

[
  {
    "process_name": "nginx",
    "group": null,
    "socket_addresses": [
      {
        "host": "0.0.0.0",
        "port": 8080,
        "is_public": true
      }
    ],
    "created": "2022-12-01T20:25:36.009608+00:00"
  }
]

# Note the host and port and concat them together
$ open http://0.0.0.0:8080

or, if you’ve got jq on macOS and want a one-liner:

open http://$(bin/hpctl list | jq -r "(.[0].socket_addresses[0].host + \":\" + (.[0].socket_addresses[0].port|tostring))")

Stopping the Process 

bin/hpctl stop nginx

Webserver (Advanced)

The file examples/nginx-advanced.json is just like the Webserver example above, but with two additions:

The process defines resource limits on both CPU and memory
The process uses node selectors to limit itself to run on a node with the label kubernetes.io/hostname=gke-hydroplane-test-clus-default-pool-ffab0dea-gmz0

Chat 

The JSON files in examples/chat are the process specs for three processes: a chat server and two chat clients. Details of how the client and server work can be found in the hydroflow repo.

Launching the Processes 

# Launch the server (from the root of this repo)
bin/hpctl start examples/chat/chat-server.json

# Give the server some time to warm up
sleep 5

# Start the clients
bin/hpctl start examples/chat/chat-client-alice.json
bin/hpctl start examples/chat/chat-client-bob.json

Making Sure It Worked 

The chat server and clients don’t have any public interfaces, so we’ll to look at their logs to make sure they’re working.

In the docker runtime:

# Check that processes exist; should see a server container and two client containers
docker ps

# Check the server's logs. It should report that it's ready.
docker logs chat-server

# Check the clients' logs. You should see both alice and bob sending random messages to each other.
docker logs chat-client-alice
docker logs chat-client-bob

In the eks runtime:

# Check that pods for each process exist; should see three of them, one per process
kubectl get pods

# Check the server's logs. It should report that it's ready.
kubectl logs pod/chat-server

# Check the clients' logs. You should see both alice and bob sending random messages to each other.
kubectl logs pod/chat-client-alice
kubectl logs pod/chat-client-bob

Stopping the Processes 

bin/hpctl stop -g chat-clients
bin/hpctl stop chat-server