[
https://issues.apache.org/jira/browse/CASSANDRA-15538?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17210626#comment-17210626
]
Alex Petrov commented on CASSANDRA-15538:
-----------------------------------------
bq. What's our appetite for taking Harry from "it is possible to do this with
Harry" to "we are in fact doing this with Harry" in the 4.0 time frame
I think we should most definitely do this. At every least, requiring
verification of a shrinking (or expanding) cluster of a modest size (up to
maybe 10 nodes, up to several gb per node) to pass for a release to qualify.
Such qualification would mean that data survives internode, compaction,
repairs, streaming.
bq. Are there areas of the codebase we feel still have significant risk that we
should further exercise with things like Harry and Fallout before release
Unfortunately areas I would feel are the most risky are also the ones that are
hardest to test: everything related to ring movement. That said, range
tombstones, upgrades, paging, SRP, reverse iteration (and combinations of
thereof) are also very important. However, this may be just recency bias: it
feels like many criticial issues reported lately were related to these
categories.
In any case, advantage of Harry is that we do not have to aim specifically at
some area of code. Of course, if there's no bootstrap, there'll be no
streaming, and if client doesn't sue paging, we're no exercising this path, but
most other things I've mentioned should be still exercised anyways.
> 4.0 quality testing: Local Read/Write Path: Other Areas
> -------------------------------------------------------
>
> Key: CASSANDRA-15538
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15538
> Project: Cassandra
> Issue Type: Task
> Components: Test/dtest/java, Test/dtest/python
> Reporter: Josh McKenzie
> Priority: Normal
> Fix For: 4.0-beta
>
>
> Reference [doc from
> NGCC|https://docs.google.com/document/d/1uhUOp7wpE9ZXNDgxoCZHejHt5SO4Qw1dArZqqsJccyQ/edit#]
> for context.
> *Shepherd: Aleksey Yeschenko*
> Testing in this area refers to the local read/write path (StorageProxy,
> ColumnFamilyStore, Memtable, SSTable reading/writing, etc). We are still
> finding numerous bugs and issues with the 3.0 storage engine rewrite
> (CASSANDRA-8099). For 4.0 we want to ensure that we thoroughly cover the
> local read/write path with techniques such as property-based testing, fuzzing
> ([example|http://cassandra.apache.org/blog/2018/10/17/finding_bugs_with_property_based_testing.html]),
> and a source audit.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
