storm-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jungtaek Lim (JIRA)" <>
Subject [jira] [Updated] (STORM-3309) TickTupleTest is still flaky
Date Sat, 16 Mar 2019 03:27:00 GMT


Jungtaek Lim updated STORM-3309:
    Fix Version/s:     (was: 2.0.1)

> TickTupleTest is still flaky
> ----------------------------
>                 Key: STORM-3309
>                 URL:
>             Project: Apache Storm
>          Issue Type: Bug
>          Components: storm-server
>    Affects Versions: 2.0.0
>            Reporter: Stig Rohde Døssing
>            Assignee: Stig Rohde Døssing
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 2.0.0
>          Time Spent: 20m
>  Remaining Estimate: 0h
> {quote}
>  testTickTupleWorksWithSystemBolt  Time elapsed: 6.802 s  <<< FAILURE!
> java.lang.AssertionError: Iteration 1 expected:<52000> but was:<51000>
> {quote}
> The test runs a topology in a local cluster with time simulation. One of the bolts has
tick tuples enabled, and the test tries to check that the ticks arrive with 1 second intervals.
> As far as I can tell, the problem is that time simulation doesn't cover the bolt and
spout executors. When the test increases simulated time by 1 second and waits for the cluster
to idle, the test expects that to mean that the bolt will at that point have consumed the
tick. In some cases this doesn't happen, and multiple tick tuples may end up queued before
the bolt consumes them. Since the bolt is responsible for generating the timestamp, the test
will fail.

This message was sent by Atlassian JIRA

View raw message