hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Raghotham Murthy (JIRA)" <>
Subject [jira] Updated: (HIVE-73) Thrift Server and Client for Hive
Date Mon, 01 Dec 2008 20:51:44 GMT


Raghotham Murthy updated HIVE-73:

    Attachment: hive-73.10.patch

> 1. printInfo vs printError 

ideally none of the console outputs other than the data itself should go to stdout. Or we
should make "hive -e" default to quiet. We shouldnt have to specify more parameters to just
get the data out. I have reverted this change for now. will file another jira for this.

> 2. has serious issues: 

I have removed these scripts. for now made a simple change to bin/hive to start the server
if 'service is provided as the first argument. note that this does not daemonize the server.

> 1. hardcoded path: +$JAVA_HOME/bin/java -Djava.library.path=/mnt/hive/production/cluster/lib/native/Linux-amd64-64/

No longer hardcoded sice i modified bin/hive

> 2. getStatus and getVersion are already defined in fb303, so why re-defining them?
> 3. usually fb303 services take advantage of set/getcounter to provide stats on #of various
ops and also exceptions. e.g., incrementCounter("queryExceptions");

I have gotten rid of fb303 daemonizing stuff until we have a clear way of just using bin/hive
for everything. i'd like the client and the server to be two different scripts. but, we could
discuss that in a separate jira.

I have disabled the failing test (should file a separate jira on serde)

> Thrift Server and Client for Hive
> ---------------------------------
>                 Key: HIVE-73
>                 URL:
>             Project: Hadoop Hive
>          Issue Type: New Feature
>          Components: Server Infrastructure
>    Affects Versions: 0.20.0
>            Reporter: Raghotham Murthy
>            Assignee: Raghotham Murthy
>             Fix For: 0.20.0
>         Attachments: hive-73.1.patch, hive-73.10.patch, hive-73.2.patch, hive-73.3.txt,
hive-73.4.txt, hive-73.5.txt, hive-73.6.patch, hive-73.7.patch, hive-73.8.patch, hive-73.9.patch
> Currently the hive cli directly calls the driver code. We need to be able to run a stand
alone hive server that multiple clients can connect to. The hive server will allow clients
to run queries as well as make meta data calls (by inheriting from the thrift metastore server)

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message