You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@nifi.apache.org by "Adam Hunyadi (Jira)" <ji...@apache.org> on 2020/09/15 13:15:00 UTC

[jira] [Created] (MINIFICPP-1373) Implement and test a simplified ConsumeKafka processor without security protocols

Adam Hunyadi created MINIFICPP-1373:
---------------------------------------

             Summary: Implement and test a simplified ConsumeKafka processor without security protocols
                 Key: MINIFICPP-1373
                 URL: https://issues.apache.org/jira/browse/MINIFICPP-1373
             Project: Apache NiFi MiNiFi C++
          Issue Type: Sub-task
    Affects Versions: 0.7.0
            Reporter: Adam Hunyadi
            Assignee: Adam Hunyadi
             Fix For: 1.0.0


*Acceptance Criteria:*

See attached test matrix plan.

Additional test (that require multiple Kafka consumers):
GIVEN two ConsumeKafkas with {color:#0747A6}different group ids{color} subscribed to the same topic
WHEN a message is published to the topic
THEN both of the ConsumeKafka processors should produce identical flowfiles

GIVEN two ConsumeKafkas with {color:#0747A6}the same group id{color} subscribed to the same topic
WHEN a message is published to the topic
THEN both of the ConsumeKafka processors should produce identical flowfiles

GIVEN two ConsumeKafkas with {color:#0747A6}the same group id{color} subscribed to the same topic with exactly two partitions with {color:#0747A6}Offset Reset{color} set to {color:#0747A6}earliest{color}.
WHEN a messages were already present on both partitions and the second one crashes
THEN the first one should process duplicates of the messages that originally came to the second (at_least_once delivery)

GIVEN two ConsumeKafkas with {color:#0747A6}the same group id{color} subscribed to the same topic with exactly two partitions with {color:#0747A6}Offset Reset{color} set to {color:#0747A6}latest{color}.
WHEN a messages were already present on both partitions and the second one crashes
THEN the first one should {color:#0747A6}not{color} process duplicates of the messages that originally came to the second (at_least_once delivery)

GIVEN two ConsumeKafkas with {color:#0747A6}the same group id{color} subscribed to the same topic with exactly two partitions with {color:#0747A6}Offset Reset{color} set to {color:#0747A6}none{color}.
WHEN a messages were already present on both partitions and the second one crashes
THEN the first one should throw an exception

*Background:*

See parent task.

*Proposal:*

This should be the first part of the implementation, the second being adding and testing multiple security protocols.



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