[
https://issues.apache.org/jira/browse/KAFKA-392?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joe Stein updated KAFKA-392:
----------------------------
Fix Version/s: (was: 0.8)
0.8.1
> Refactor and optimize system tests
> ----------------------------------
>
> Key: KAFKA-392
> URL: https://issues.apache.org/jira/browse/KAFKA-392
> Project: Kafka
> Issue Type: Sub-task
> Components: core
> Reporter: Joel Koshy
> Priority: Minor
> Labels: replication-testing
> Fix For: 0.8.1
>
>
> It would be helpful to refactor the system tests to make them as concise and
> clear as possible, and get them to complete faster. Obviously this is
> non-urgent but should ideally be done prior to adding new system tests if
> any.
> 1 - Several functions (e.g., wait_for_zero_consumer_lag, start_cluster,
> etc.) are all util methods that are relevant across system tests. Also,
> these util functions can be switched to getopt (preferrably with long
> options) which will make the scripts more readable. Anything that is not
> super-specific to any test should be moved to the global util script.
> 2 - Might help to have all test output and logs sent to a directory in tmp
> (use mktemp) so everything is contained there - right now some files are
> local and some are in tmp. Also, clean up would simply be rm -rf the temp
> directory instead of handling multiple locations.
> 3 - We should get rid of as many sleeps as possible. E.g., with producer
> acks I don't think we need to sleep after producer-performance. Likewise
> for all other sleeps, let us see if they are really needed and provide
> tooling (if necessary) to eliminate/reduce them.
> 4 - User-variable test parameters should be command-line options (getopt) to
> the script.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
