phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Geoffrey Jacoby (JIRA)" <>
Subject [jira] [Commented] (PHOENIX-3983) Index rebuild scans should not be using the ServerRpcControllerFactory
Date Fri, 30 Jun 2017 21:14:02 GMT


Geoffrey Jacoby commented on PHOENIX-3983:

+1 to a separate HBase 1.4 code line (after HBASE-15816 is committed to branch-1), since hopefully
that will also encourage the release of HBase 1.4. I have JIRAs that have been sitting in
there for a year. :-) 

The problem with using a branch-1.3 version of HBASE-15816 in Phoenix is that then Phoenix
master, or a future 4.x-HBase-1.3 branch is actually depending on a future HBase 1.3.2, not
1.3, which might be difficult to communicate to operators trying to use vanilla Apache distributions
of one or both. (The major distros could just package the "right" two versions together and
hide the complexity for their customers.) Unless there are already things in HBase 1.3.1 but
not 1.3 that Phoenix won't compile without? 

Wrt the options [~vincentpoon] lists, my ordered preference would be:
1. Separate 4.x-HBase-1.4 ( or 1.3) branch
2. Injected PrioritySetter as a stopgap
3. Lots of ugly instanceofs or reflection method checks as a stopgap. 

> Index rebuild scans should not be using the ServerRpcControllerFactory
> ----------------------------------------------------------------------
>                 Key: PHOENIX-3983
>                 URL:
>             Project: Phoenix
>          Issue Type: Bug
>            Reporter: Samarth Jain
>            Assignee: Samarth Jain
>         Attachments: PHOENIX-3983.patch

This message was sent by Atlassian JIRA

View raw message