phoenix-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Talanki, Vikash" <>
Subject Re: Phoenix 4.7 CSVBulk Loading not populating index tables
Date Fri, 17 Jun 2016 16:43:04 GMT
Hi Gabriel,

Yes, your understanding on latest code change for CSVBUlkLoad is correct - run 1 MR job and
populate data and all index tables.
I see that its running one MR job but populating data table and only the first index table
that has been created on that table.

However, I came to know that this bug exists in 4.7 but is resolved as part of 4.8(yet to
be released). I have taken the latest patch and used it and seems like this issue is resolved
Its running a single map reduce job and populating all index tables.

Vikash Talanki

On 6/17/16, 8:30 AM, "Gabriel Reid" <> wrote:

>Hi Vikash,
>If I'm not mistaken, the bulk load tool was changed in 4.7 to populate
>the main table and index tables in a single job (instead of one job
>per table).
>However, based on what you're seeing, it sounds like there's a problem
>with this change.
>Could you verify that only one index table was populated, and if so,
>log a bug in Jira at ?
>On Tue, Jun 14, 2016 at 8:18 PM, Talanki, Vikash <> wrote:
>> Hi All,
>> I'm currently trying to load CSV files in Phoenix table using CSVBulkLoad
>> tool using phoenix-4.7.0-HBase-1.1-client.jar in Hortonworks.
>> I have a phoenix table and 6 indexes were created on this table. According
>> to CSV Bulk Load documentation -
>>, it should run 7 MR jobs and
>> load data into main table and 6 index tables.
>> But I'm seeing that when I use the above 4.7 jar, it is running only 1 MR
>> job and loading data into only main data table and first index table.
>> When I use 4.5.2 jar, it runs 7 MR jobs as documented and loads the data in
>> all tables.
>> So is this the issue with phoenix-4.7.0-HBase-1.1-client.jar jar? Any work
>> around if I want to use 4.7 jar?
>> Thanks,
>> Vikash Talanki
View raw message