hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nicolas Liochon (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-9719) Premptive Call Me Maybe HBase
Date Thu, 11 Sep 2014 08:00:42 GMT

    [ https://issues.apache.org/jira/browse/HBASE-9719?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14129770#comment-14129770

Nicolas Liochon commented on HBASE-9719:

There is this: "java.lang.OutOfMemoryError: unable to create new native thread"
Does jepsen creates a new thread per client call? If so that's a lot of threads (and as hbase
retries for around 10 minutes by default, you can have a lot of threads running simultaneously).
If you change the number of retries to something like 3 or 4 then the ops will fail sooner.

> Premptive Call Me Maybe HBase
> -----------------------------
>                 Key: HBASE-9719
>                 URL: https://issues.apache.org/jira/browse/HBASE-9719
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: Robert Yokota
>             Fix For: 0.98.1
> Aphyr wrote an interesting article on C* [1].  Some awkward-looking issues were turned
up though it seems the author is purportedly doing nothing but exercising the software within
spec; he is just paying close attention to what is being returned.
> It does not look like Aphyr will be coming our way any time soon [2] -- thanks Ian Varley
-- but he could change his mind.  Wouldn't it be coolio if we'd already run his test suite
and found any bugs and fixed them before he came by?  This issue is about running his article
against hbase so we find the embarrassing before he does.
> 1. http://aphyr.com/posts/294-call-me-maybe-cassandra
> 2. https://twitter.com/aphyr/status/335082835868254209

This message was sent by Atlassian JIRA

View raw message