[jira] [Comment Edited] (CASSANDRA-7371) DELETEs get lost

["Robert Stupp (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-7371?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14025380#comment-14025380 ] 

Robert Stupp edited comment on CASSANDRA-7371 at 6/9/14 6:45 PM:
-----------------------------------------------------------------

Unit test attached.

Instructions:
# start local cluster ({{ccm create -v 2.1.0-rc1-snapshot -n 1 -s --vnodes test}} - or manually)
# start the attached unit test (dependency to java driver (2.0 should work - tested with latest 2.1 from github))

* Works with C* build against SHA 78e91c4cacb7128a994a7a8ca16f5a9e20b576ce
* Some iterations fail with C* build against SHA 4722fe70aa9ae1b62772cfa1a1de58ef289445d5



was (Author: snazy):
Unit test attached.

Instructions:
# start local cluster ({{ccm create -v 2.1.0-rc1-snapshot -n 1 -s --vnodes test}} - or manually)
# start the attached unit test (dependency to java driver (2.0 should work - tested with latest 2.1 from github))

* Works with C* build against SHA 60be74230e48bb7f3aba132e247d054a8b8b8a0d
* Some iterations fail with C* build against SHA 029cd403172f8cc1b96f0828c8e1f8f6d57c0bcc


> DELETEs get lost
> ----------------
>
>                 Key: CASSANDRA-7371
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7371
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>         Environment: 2.1 git branch since merge commit 4722fe70aa9ae1b62772cfa1a1de58ef289445d5 ("RefCount native frames from netty to avoid corruption bugs")
>            Reporter: Robert Stupp
>            Assignee: Aleksey Yeschenko
>            Priority: Blocker
>             Fix For: 2.1.0
>
>         Attachments: Cassandra7371.java
>
>
> The mentioned commit (merged to 2.1 from 1.2 via 2.0) introduced a bug which is not easy to reproduce:
> Workload description:
> - One INSERT into a table
> - multiple concurrent SELECTs against different tables (one select returns a result)
> - One UPDATE against the same table as the INSERT
> - (same) multiple concurrent SELECTs against different tables (one select returns a result)
> - One DELETE against the same table as the INSERT
> - (same) multiple concurrent SELECTs against different tables
> Expected is that the last bunch of SELECTs returns no result. But since commit SHA  the DELETE gets not processed.
> To clarify - the DELETE is not delayed - it is not executed at all.
> Checked against a single node C* "cluster".
> The above works with 2.0.8 and against 2.1.0-* - but not C* code since git commit 4722fe70aa9ae1b62772cfa1a1de58ef289445d5 ("RefCount native frames from netty to avoid corruption bugs") from June, 5th.
> I am not sure whether this bug also occurs in 1.2 or 2.0 branches (didn't check that).
> It is not always reproducible - approx 1 of 50 tries run into this issue.
> If you have difficulties in finding the reason, I can provide a unit test for that.



--
This message was sent by Atlassian JIRA
(v6.2#6252)