[
https://issues.apache.org/jira/browse/NIFI-5044?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16453025#comment-16453025
]
Matt Burgess commented on NIFI-5044:
------------------------------------
Here are my thoughts on non-happy-path situations:
If the Pre-Query has no EL and fails, the user configured the processor
"incorrectly"; no flow files will be processed successfully. That's usually
when we yield and rollback, saying "hey this thing is broken". Pre-Query
failures with EL it might be the flow file's fault, so failure is probably the
right way to go. However we may want to penalize it so if it gets re-routed
back to the processor (like a fake "retry") it won't hammer the processor when
the query is probably going to fail all the time for that FF.
Post-Query failures may have the same features, but the query has finished
successfully so we might as well make the data available, along with any
information about Post-Query failures (for example, if you were expecting an
MSCK to succeed after the fact and it didn't).
So I guess my take is: 1) Yield & Rollback on Pre-Query failure with no EL, 2)
Route to Failure (with penalization) for everything else. If we know more about
the different kinds of exceptions that can be produced, perhaps we can do more
finer-grained things. We could also add a failure.reason attribute to the
failed FFs so people could do routing accordingly (and perhaps failure.code if
there's something like a JDBC return code or something associated with the
status).
> SelectHiveQL accept only one statement
> --------------------------------------
>
> Key: NIFI-5044
> URL: https://issues.apache.org/jira/browse/NIFI-5044
> Project: Apache NiFi
> Issue Type: Improvement
> Affects Versions: 1.2.0
> Reporter: Davide Isoardi
> Priority: Critical
>
> In [this
> |[https://github.com/apache/nifi/commit/bbc714e73ba245de7bc32fd9958667c847101f7d]
> ] commit claims to add support to running multiple statements both on
> SelectHiveQL and PutHiveQL; instead, it adds only the support to PutHiveQL,
> so SelectHiveQL still lacks this important feature. @Matt Burgess, I saw that
> you worked on that, is there any reason for this? If not, can we support it?
> If I try to execute this query:
> {quote}set hive.vectorized.execution.enabled = false; SELECT * FROM table_name
> {quote}
> I have this error:
>
> {quote}2018-04-05 13:35:40,572 ERROR [Timer-Driven Process Thread-146]
> o.a.nifi.processors.hive.SelectHiveQL
> SelectHiveQL[id=243d4c17-b1fe-14af-ffff-ffffee8ce15e] Unable to execute
> HiveQL select query set hive.vectorized.execution.enabled = false; SELECT *
> FROM table_name for
> StandardFlowFileRecord[uuid=0e035558-07ce-473b-b0d4-ac00b8b1df93,claim=StandardContentClaim
> [resourceClaim=StandardResourceClaim[id=1522824912161-2753,
> container=default, section=705], offset=838441,
> length=25],offset=0,name=cliente_attributi.csv,size=25] due to
> org.apache.nifi.processor.exception.ProcessException: java.sql.SQLException:
> The query did not generate a result set!; routing to failure: {}
> org.apache.nifi.processor.exception.ProcessException: java.sql.SQLException:
> The query did not generate a result set!
> at
> org.apache.nifi.processors.hive.SelectHiveQL$2.process(SelectHiveQL.java:305)
> at
> org.apache.nifi.controller.repository.StandardProcessSession.write(StandardProcessSession.java:2529)
> at
> org.apache.nifi.processors.hive.SelectHiveQL.onTrigger(SelectHiveQL.java:275)
> at
> org.apache.nifi.processors.hive.SelectHiveQL.lambda$onTrigger$0(SelectHiveQL.java:215)
> at
> org.apache.nifi.processor.util.pattern.PartialFunctions.onTrigger(PartialFunctions.java:114)
> at
> org.apache.nifi.processor.util.pattern.PartialFunctions.onTrigger(PartialFunctions.java:106)
> at
> org.apache.nifi.processors.hive.SelectHiveQL.onTrigger(SelectHiveQL.java:215)
> at
> org.apache.nifi.controller.StandardProcessorNode.onTrigger(StandardProcessorNode.java:1120)
> at
> org.apache.nifi.controller.tasks.ContinuallyRunProcessorTask.call(ContinuallyRunProcessorTask.java:147)
> at
> org.apache.nifi.controller.tasks.ContinuallyRunProcessorTask.call(ContinuallyRunProcessorTask.java:47)
> at
> org.apache.nifi.controller.scheduling.TimerDrivenSchedulingAgent$1.run(TimerDrivenSchedulingAgent.java:132)
> at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308)
> at
> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180)
> at
> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: java.sql.SQLException: The query did not generate a result set!
> at org.apache.hive.jdbc.HiveStatement.executeQuery(HiveStatement.java:438)
> at
> org.apache.commons.dbcp.DelegatingStatement.executeQuery(DelegatingStatement.java:208)
> at
> org.apache.commons.dbcp.DelegatingStatement.executeQuery(DelegatingStatement.java:208)
> at
> org.apache.nifi.processors.hive.SelectHiveQL$2.process(SelectHiveQL.java:293)
> {quote}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
