You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafodion.apache.org by "Hans Zeller (JIRA)" <ji...@apache.org> on 2016/05/20 01:20:16 UTC
[jira] [Updated] (TRAFODION-1014) LP Bug: 1421747 - SQL Upsert
using load periodically not saving all rows
[ https://issues.apache.org/jira/browse/TRAFODION-1014?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Hans Zeller updated TRAFODION-1014:
-----------------------------------
Fix Version/s: (was: 2.0-incubating)
2.1-incubating
> LP Bug: 1421747 - SQL Upsert using load periodically not saving all rows
> ------------------------------------------------------------------------
>
> Key: TRAFODION-1014
> URL: https://issues.apache.org/jira/browse/TRAFODION-1014
> Project: Apache Trafodion
> Issue Type: Bug
> Components: sql-cmp
> Reporter: Gary W Hall
> Assignee: Suresh Subbiah
> Priority: Critical
> Fix For: 2.1-incubating
>
>
> When running a script that initiates 32 parallel streams loading a table, we have found that periodically there are gaps in the resulting saved data...for example we will find that we are missing stock items #29485 thru #30847 inclusive for Warehouse #5. The number of gaps found for a given load run varies...normally none, but I've seen as many as eight gaps of missing data.
> The sql statement used in all streams is as follows:
> sql_statement = "upsert using load into " + stock_table_name
> + " (S_I_ID, S_W_ID, S_QUANTITY, S_DIST_01, S_DIST_02, S_DIST_03, S_DIST_04,"
> + " S_DIST_05, S_DIST_06, S_DIST_07, S_DIST_08, S_DIST_09, S_DIST_10,"
> + " S_YTD, S_ORDER_CNT, S_REMOTE_CNT, S_DATA)"
> + " values (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)";
> This is not easily repeatable…I’ve run the script to drop/create/load this table 12 times today, resulting in some missing rows 4 of the 12 times. Worst case we were missing 0.03% of the required rows in the table…obviously, ANY missing data is not acceptable.
> Our test environment control parameters (in case any are of value to you)...
> OrderEntryLoader
> Load Starting : 2015-02-13 04:58:13
> PropertyFile : trafodion.properties
> Datebase : trafodion
> Schema : trafodion.javabench
> ScaleFactor : 512
> Streams : 32
> Maintian : true
> Load : true
> AutoCommit : true
> BatchSize : 1000
> Upsert : true
> UsingLoad : true
> IntervalLength : 60
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)