Skip to content

YarnJobsMonitor use new DateTime rather than getting system time #438

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
blbarker wants to merge 1 commit into
base: master
Choose a base branch
from
Open

YarnJobsMonitor use new DateTime rather than getting system time #438

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 16 additions & 5 deletions ...ne-core/src/main/scala/org/trustedanalytics/atk/engine/command/mgmt/YarnJobsMonitor.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@ package org.trustedanalytics.atk.engine.command.mgmt

import java.util.concurrent.TimeUnit

import org.joda.time.DateTime
import org.trustedanalytics.atk.domain.jobcontext.JobContext
import org.trustedanalytics.atk.engine.{ EngineConfig, Engine }
import org.trustedanalytics.atk.engine.plugin.Invocation
Expand All @@ -34,16 +35,26 @@ class YarnJobsMonitor(engine: Engine)(implicit invocation: Invocation) extends R
while (true) {
engine.getCommandsNotComplete().foreach { command =>
engine.getCommandJobContext(command) match {
case Some(context) => if (hasStaleContext(context)) {
engine.cancelCommand(command.id, Some(s" by ATK context monitor due to timeout. The job context ${context.clientId} has not provided an update for more than $timeoutMinutes minutes. This may indicate that a task is running for a very long time. Try increasing the 'trustedanalytics.atk.engine.yarn-monitor-task-timeout' config setting."))
}
case Some(context) =>
val (isStale, msg) = hasStaleContext(context)
if (isStale) {
engine.cancelCommand(command.id, Some(s" by ATK context monitor due to timeout. The job context ${context.clientId} has not provided an update for more than $timeoutMinutes minutes. This may indicate that a task is running for a very long time. Try increasing the 'trustedanalytics.atk.engine.yarn-monitor-task-timeout' config setting. Details: $msg"))
}
case None => ; // there is no know YARN job to shutdown (command remains not complete, but this is not the responsibility of a YARN jobs monitor
}
}
TimeUnit.MINUTES.sleep(timeoutMinutes)
}
}

def hasStaleContext(context: JobContext): Boolean =
System.currentTimeMillis() - context.modifiedOn.getMillis > timeoutMinutes * 60 * 1000
def hasStaleContext(context: JobContext): (Boolean, String) = {
val modMillis = context.modifiedOn.getMillis
val nowMillis = new DateTime().getMillis
val timeoutMillis = timeoutMinutes * 60 * 1000
val answer = nowMillis - modMillis > timeoutMillis
val msg = s"YarnJobsMonitor hasStaleContext: $nowMillis - $modMillis > $timeoutMillis is $answer"
info(msg)
(answer, msg)
//System.currentTimeMillis() - context.modifiedOn.getMillis > timeoutMinutes * 60 * 1000
}
}