[
https://issues.apache.org/jira/browse/HAWQ-1426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15960282#comment-15960282
]
Lili Ma edited comment on HAWQ-1426 at 4/7/17 5:12 AM:
-------------------------------------------------------
RCA:
When hawq extract tries to find the HDFS files information, it wrongly treated
the pg_aoseg.pg_aoseg_$relid as the catalog table for storing those
information.
Here is the correct processing steps:
1. Find the directory on HDFS which stores the actual data for the table. This
can be achieved by following the column "relfilenode" in pg_class table.
2. Find the detailed file name for the table under above directory. This can be
achieved by searching the catalog table pg_aoseg.pg_aoseg(paqseg)_$. The table
name suffix is neither $relid nor $relfilenode under some circumstances. We
should get it by referring the column "segrelid" in catalog table
"pg_appendonly", and then looking up the table "pg_class" to get the accurate
table name.
was (Author: lilima):
RCA:
When hawq extract tries to find the HDFS files information, it wrongly treated
the pg_aoseg.pg_aoseg_$relid as the catalog table for storing those
information.
When determining the file path of a table, hawq extract should follow below
steps:
1. Find the directory on HDFS which stores the actual data for the table. This
can be achieved by following the column "relfilenode" in pg_class table.
2. Find the detailed file name for the table under above directory. This can be
achieved by searching the catalog table pg_aoseg.pg_aoseg(paqseg)_$. The table
name suffix is neither $relid nor $relfilenode under some circumstances. We
should get it by referring the column "segrelid" in catalog table
"pg_appendonly", and then looking up the table "pg_class" to get the accurate
table name.
> hawq extract meets error after the table was reorganized.
> ---------------------------------------------------------
>
> Key: HAWQ-1426
> URL: https://issues.apache.org/jira/browse/HAWQ-1426
> Project: Apache HAWQ
> Issue Type: Bug
> Components: Command Line Tools
> Reporter: Lili Ma
> Assignee: Ed Espino
> Fix For: 2.3.0.0-incubating
>
>
> After one table is reorganized, hawq extract the table will meet error.
> Reproduce Steps:
> 1. create an AO table
> 2. insert into several records into it
> 3. Get the table reorganized. "alter table a set with (reorganize=true);"
> 4. run hawq extract, error thrown out.
> For the bug fix, we should also guarantee that hawq extract should work if
> the table is truncated and re-inserted.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
