Barry G Becker

If the jobId was included in the typesafe configuration that was passed to the SparkSqlJob.runJob method, then I could keep track of job progress within my own app and make...

Cool. It worked, though I'm not sure why it calls the new API instead of the old. In my class I have: ``` object MyJob extends SparkSqlJob { override def...

I would really like to avoid going back to the JobFileDao, but the JobSqlDao seems to be less reliable in our cucumber bdd testing. For example, when using the JobSqlDao...

@ecandreev, I know because I first wait for the job to be finished (i.e status != RUNNING) with code like this in my RestSteps.scala (using cucumber-scala from templemore): ``` def...

@velvia, I'm not sure exactly what you mean by sql config, but here is the sqlcontext section in the local.conf file: ``` sqldao { slick-driver = slick.driver.MySQLDriver jdbc-driver = com.mysql.jdbc.Driver...

Yes, we need to use sync because a) we want to show progress for long running jobs, and b) a long sync request will trigger timeouts in jobserver. I don't...

I don't see how. The isJobDone method only returns when the status in the response for the specified jobId is status != "RUNNING". Then when I go to check the...

In isJobDone above, at the point when status != "RUNNING" I now print the status and the result. What I am seeing in the cases when isJobDone returns true and...

Not sure what you would like to see when you ask for the function that is making the async request. We make an http POST request through java (using org.springframework.web.client.RestTemplate),...

I also noticed that international characters returned in json were coming through corrupted (appeared as question marks inside of diamonds). I'm not sure, but I think this line in spark/jobserver/ChunkEncodingActor...