[jira] [Created] (FLINK-16488) Logging issues when running through K8s

[Shang Yuanchun (Jira)]

John Mathews created FLINK-16488:
------------------------------------

             Summary: Logging issues when running through K8s
                 Key: FLINK-16488
                 URL: https://issues.apache.org/jira/browse/FLINK-16488
             Project: Flink
          Issue Type: Bug
          Components: Deployment / Kubernetes
    Affects Versions: 1.8.0
         Environment: I am running with the K8's environment described here:

[https://ci.apache.org/projects/flink/flink-docs-stable/ops/deployment/kubernetes.html]

 

Here is a sample of the code I am running with: 
{code:java}
public static void main(String[] args) {
    log.info("_TOKEN_LOG_1");
    System.out.println("_TOKEN_SOUT_1");
    // ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
    ParameterTool params = ParameterTool.fromArgs(args);
    final ExecutionEnvironment env = ExecutionEnvironment.getExecutionEnvironment();
    log.info("_TOKEN_LOG_2");
    System.out.println("_TOKEN_SOUT_1");
    // make parameters available in the web interface
    env.getConfig().setGlobalJobParameters(params);

    // get default test text data
    DataSet<String> text = getTextDataSet(env);
    log.info("_TOKEN_LOG_3");
    System.out.println("_TOKEN_SOUT_1");
    DataSet<Tuple2<String, Integer>> counts =
            // split up the lines in pairs (2-tuples) containing: (word,1)
            text.flatMap(new Tokenizer())
                    // group by the tuple field "0" and sum up tuple field "1"
                    .groupBy(0)
                    .sum(1);
    log.info("_TOKEN_LOG_4");
    System.out.println("_TOKEN_SOUT_1");
    // emit result
    if (params.has("output")) {
        counts.writeAsCsv(params.get("output"), "\n", " ");
        // execute program
        try {
            env.execute("WordCount Example");
        } catch (Exception e) {
            e.printStackTrace();
            log.info("_TOKEN_LOG_ERROR", e);
            System.out.println("_TOKEN_SOUT_ERROR" + e.toString());
        }
    } else {
        log.info("_TOKEN_LOG_5");
        System.out.println("_TOKEN_SOUT_1");
        try {
            counts.print();
        } catch (Exception e) {
            e.printStackTrace();
            log.info("_TOKEN_LOG_ERROR2", e);
            System.out.println("_TOKEN_SOUT_ERROR2" + e.toString());
        }
        log.info("_TOKEN_LOG_6");
        System.out.println("_TOKEN_SOUT_1");
    }
}

private static DataSet<String> getTextDataSet(ExecutionEnvironment env) {
    // get default test text data
    return env.fromElements(
            "To be, or not to be,--that is the question:--",
            "Whether 'tis nobler in the mind to suffer",
            "The slings and arrows of outrageous fortune",
            "Or to take arms against a sea of troubles,",
            "And by opposing end them?--To die,--to sleep,--",
            "No more; and by a sleep to say we end",
            "The heartache, and the thousand natural shocks",
            "That flesh is heir to,--'tis a consummation",
            "Devoutly to be wish'd. To die,--to sleep;--",
            "To sleep! perchance to dream:--ay, there's the rub;",
            "For in that sleep of death what dreams may come,",
            "When we have shuffled off this mortal coil,",
            "Must give us pause: there's the respect",
            "That makes calamity of so long life;",
            "For who would bear the whips and scorns of time,",
            "The oppressor's wrong, the proud man's contumely,",
            "The pangs of despis'd love, the law's delay,",
            "The insolence of office, and the spurns",
            "That patient merit of the unworthy takes,",
            "When he himself might his quietus make",
            "With a bare bodkin? who would these fardels bear,",
            "To grunt and sweat under a weary life,",
            "But that the dread of something after death,--",
            "The undiscover'd country, from whose bourn",
            "No traveller returns,--puzzles the will,",
            "And makes us rather bear those ills we have",
            "Than fly to others that we know not of?",
            "Thus conscience does make cowards of us all;",
            "And thus the native hue of resolution",
            "Is sicklied o'er with the pale cast of thought;",
            "And enterprises of great pith and moment,",
            "With this regard, their currents turn awry,",
            "And lose the name of action.--Soft you now!",
            "The fair Ophelia!--Nymph, in thy orisons",
            "Be all my sins remember'd.");
}
{code}
 
            Reporter: John Mathews


When running a slim downed version of the wordcount example, I am seeing some very strange logging behavior when running using the K8's setup described on the site. Essentially, every log line before the env.execute() shows up and every log line after does not (verified both through the UI, and by grepping within the container itself through K8s). 

 

Running the code below displays the following output: 

 
2020-03-07 23:14:06,967 INFO _TOKEN_LOG_1
2020-03-07 23:14:06,968 INFO _TOKEN_LOG_2
2020-03-07 23:14:06,968 INFO _TOKEN_LOG_3
2020-03-07 23:14:06,970 INFO _TOKEN_LOG_4
2020-03-07 23:14:06,970 INFO _TOKEN_LOG_5
 

The job completes successfully, but there is no sign of either a _TOKEN_LOG_6 nor a _TOKEN_LOG_ERROR. I have tested a few permutations of this, and all logging stops as soon as an environment.execute(*) command is called.

 

 

Any idea on what is happening to these logs?



--
This message was sent by Atlassian Jira
(v8.3.4#803005)