You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@samza.apache.org by "Vishal Kuo (JIRA)" <ji...@apache.org> on 2016/04/15 04:46:25 UTC

[jira] [Created] (SAMZA-935) hello-samza add kafka wait for dependent services

Vishal Kuo created SAMZA-935:
--------------------------------

             Summary: hello-samza add kafka wait for dependent services
                 Key: SAMZA-935
                 URL: https://issues.apache.org/jira/browse/SAMZA-935
             Project: Samza
          Issue Type: Improvement
          Components: hello-samza
            Reporter: Vishal Kuo
            Priority: Minor


When working with a modified bin/grid script found in https://github.com/theduderog/hello-samza-confluent we found that attempting to start confluent's schemaregistry service immediately after starting the other services resulted in failure since schemaregistry is dependent on kafka which takes a couple seconds to start up.

Would it make sense to add a wait option for kafka like we have added here: https://github.com/theduderog/hello-samza-confluent/blob/master/bin/grid#L118-L132 ? 
```
wait_kafka() {
  echo "Waiting for kafka broker to start..."
  until $(nc -w 2 localhost 9092); do
    printf '.'
    sleep 1
  done
  printf '\n'
  echo "Kafka broker is listening";
}
```
This way, dependent services started right after ./bin/grid bootsrap won't fail intermittently. 

I don't mind adding this if it can be considered useful.




--
This message was sent by Atlassian JIRA
(v6.3.4#6332)