[
https://issues.apache.org/jira/browse/CASSANDRA-20148?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17916094#comment-17916094
]
Branimir Lambov commented on CASSANDRA-20148:
---------------------------------------------
I did not find a better way to fix this. The difference comes from the limiter
starting with 1s of accumulated time. I tried to clear that by changing the
rate as well as by running another compaction, enough time still passes between
calls to get that one second buffer.
I also tried to find a way to take the duration of all three sstables'
compactions, but {{watch_log_for}} repeatedly returns the first entry that
matches even if you query for multiple {{Compacted}} lines.
In the end I'm just giving it more leniency, calculated from the expected
duration of the op. I'm also taking advantage of this opportunity to add
{{UnifiedCompactionStrategy}} to the strategies tested by
{{{}compaction_test{}}}.
> Test failure: compaction_test.py::TestCompaction::test_compaction_throughput
> -----------------------------------------------------------------------------
>
> Key: CASSANDRA-20148
> URL: https://issues.apache.org/jira/browse/CASSANDRA-20148
> Project: Apache Cassandra
> Issue Type: Bug
> Components: Test/dtest/python
> Reporter: Sam Tunnicliffe
> Priority: Normal
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Seeing this fail consistently on trunk with:
> {code}
> > assert float(threshold) + 0.5 >= avgthroughput_mb
> E AssertionError: assert (5.0 + 0.5) >= 5.642
> E + where 5.0 = float('5')
> compaction_test.py:302: AssertionError
> {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
