HeartSaVioR commented on a change in pull request #26821:
[SPARK-20656][CORE]Support Incremental parsing of event logs in SHS
URL: https://github.com/apache/spark/pull/26821#discussion_r357034401
##########
File path:
core/src/test/scala/org/apache/spark/deploy/history/FsHistoryProviderSuite.scala
##########
@@ -190,6 +191,54 @@ class FsHistoryProviderSuite extends SparkFunSuite with
Matchers with Logging {
provider.mergeApplicationListingCall should be (1)
}
+ test("support incremental parsing of the event logs") {
+ val provider = new FsHistoryProvider(createTestConf(true))
+
+ var store: InMemoryStore = null
+ val logFile1 = newLogFile("app1", None, inProgress = true)
+ writeFile(logFile1, None,
+ SparkListenerApplicationStart("app1", Some("app1"), 1L, "test", None),
+ SparkListenerJobStart(0, 2L, Seq())
+ )
+ updateAndCheck(provider) { list =>
+ list.size should be (1)
+ provider.getAttempt("app1", None).logPath should
endWith(EventLogFileWriter.IN_PROGRESS)
+ val appUi = provider.getAppUI("app1", None)
+ appUi should not be null
+ store = appUi.get.ui.store.store.asInstanceOf[InMemoryStore]
+ }
+
+ writeFile(logFile1, None,
+ SparkListenerApplicationStart("app1", Some("app1"), 1L, "test", None),
+ SparkListenerJobStart(0, 2L, Seq()),
+ SparkListenerJobEnd(0, 3L, JobSucceeded)
+ )
+
+ updateAndCheck(provider) { list =>
+ store should not be null
+ store.read(classOf[AppStatusListenerData], Array(Some("app1"), None))
should not be null
+ list.size should be (1)
+ provider.getAttempt("app1", None).logPath should
endWith(EventLogFileWriter.IN_PROGRESS)
+ val appUi = provider.getAppUI("app1", None)
+ appUi should not be null
Review comment:
I think it should really check whether store reflects the events correctly.
It's not enough to just check it's loaded or not. That should cover these cases
- initial read / new addition of events in same file / new file.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
