accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Billie Rinaldi (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ACCUMULO-898) look into replacing cloudtrace
Date Tue, 21 Oct 2014 20:38:34 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14179050#comment-14179050
] 

Billie Rinaldi commented on ACCUMULO-898:
-----------------------------------------

I'm working on a major refactoring of this patch that attempts to address people's feedback,
including Christopher's suggestion that we shouldn't depend on the hadoop configuration of
span receivers.  I'm hoping to post the new patch this week.  So, don't spend too much time
with the old patch.  But I'm guessing the failure you're seeing could be due to the renaming
of the tracing property to "hadoop.htrace.spanreceiver.classes" in hadoop.  Check AbstractMacIT.configureForTracing.

> look into replacing cloudtrace
> ------------------------------
>
>                 Key: ACCUMULO-898
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-898
>             Project: Accumulo
>          Issue Type: Improvement
>            Reporter: Eric Newton
>            Assignee: James Fiori
>             Fix For: 1.7.0
>
>         Attachments: ACCUMULO-898-1.patch, Cloudtrace Replacement Design Doc.pdf, trace-tracer.png,
trace-zipkin.png
>
>
> HBase has created their own distributed tracing library, and today I bumped into zipkin.
 zipkin has a reasonable visualization, and seems to work with thrift.  We should look into
replacing our tracing with one of these.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message