incubator-bigtop-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Bayer (Updated) (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (BIGTOP-520) HBase Debian builds run do-component-build twice
Date Tue, 10 Apr 2012 14:59:16 GMT

     [ https://issues.apache.org/jira/browse/BIGTOP-520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Andrew Bayer updated BIGTOP-520:
--------------------------------

    Description: HBase's debian rules has a build target but no build-stamp target. Hadoop,
on the other hand, has both. So Hadoop's build is smart enough to know it only needs to call
the build target once, since build-stamp is satisfied by the file it touches after running
do-component-build, while HBase ends up running the full build target again for later dependencies,
because there's no way for it to know it's already run.  (was: Some of the Debian builds (e.g.,
HBase, but not Hadoop) have a build target but no build-stamp target. Hadoop, on the other
hand, has both. So Hadoop's build is smart enough to know it only needs to call the build
target once, since build-stamp is satisfied by the file it touches after running do-component-build,
while HBase ends up running the full build target again for later dependencies, because there's
no way for it to know it's already run.)
        Summary: HBase Debian builds run do-component-build twice  (was: Some Debian builds
run do-component-build twice)

Actually, this is unique to HBase.
                
> HBase Debian builds run do-component-build twice
> ------------------------------------------------
>
>                 Key: BIGTOP-520
>                 URL: https://issues.apache.org/jira/browse/BIGTOP-520
>             Project: Bigtop
>          Issue Type: Bug
>          Components: Debian
>    Affects Versions: 0.3.0
>            Reporter: Andrew Bayer
>            Assignee: Andrew Bayer
>
> HBase's debian rules has a build target but no build-stamp target. Hadoop, on the other
hand, has both. So Hadoop's build is smart enough to know it only needs to call the build
target once, since build-stamp is satisfied by the file it touches after running do-component-build,
while HBase ends up running the full build target again for later dependencies, because there's
no way for it to know it's already run.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message