Skip Ribbon Commands
Skip to main content
Navigate Up
Sign In

Quick Launch

Average Rating:

facebook Twitter
Email
Print Bookmark Alert me when this article is updated

Feedback

IIR Synchronizer does not synchronize records when started separately for multiple IDTs in IIR 9.5 HotFix 1
Problem Description
In IIR 9.5 HotFix 1, the Synchronizer is not synchronizing records when started separately for multiple IDTs. 

Example

There are two IDTs, IDT_1 and IDT_2:
  • IDT_1 reads data from table1
  • IDT_2 reads records from table2
When table1 is updated first with 1000 records, in the TXN table there will be records from SEQUENCE number 1 to 1000 for IDT_1.  If table2 is updated with 200 records, in the TXN table there will be records from SEQUENCE number 1001 to 1200 for IDT_2.
 
Now, when Synchronizer is started the following is observed:
  • If the Synchronizer is started for both IDT_1 and IDT_2 separately, IDT_2 will not be updated unless Synchronizer has processed all the records for IDT_1 (i.e. 1000 records). Once it is complete, it will process records for IDT_2.
  • If the Synchronizer is started for IDT_1 alone, it runs fine and updates all the records. However, if Synchronizer is started for IDT_2, it does not process any records and does nothing.
The log files show, ideal duty cycles for the Synchronizer of IDT_2.
Cause
Solution
This is a known issue and IIR-1481 has been created for this issue to be addressed in a future release.

Workaround

 
To resolve this issue, do as follows:
  • For the IDT having lesser update, start the Synchronizer with a very high commit rate (5000 or 10000) so that it picks up the record related to the IDT. If the commit rate (-mRate parameter) is set to a high value, it will behave in the desired way.
  • For the other IDT, start the Synchronizer normally or with less commit rate. 
More Information

When the Update Synchronizer runs, it will read up to the rate value processing any of the transactions that match the idt name specified. At the end of the duty cycle (after reading rate records), it will do a commit if any records were processed.

 

By setting a low sleep value, it will process the next batch more quickly so that even if no records were present in the last duty cycle, there will be no long wait before it processes the next batch.

Applies To
Product: Identity Resolution
Problem Type:
User Type:
Project Phase:
Product Version: Identity Resolution 9.5 HotFix; Identity Resolution (Identity Search Server)
Database:
Operating System:
Other Software:

Reference
Attachments
Last Modified Date:6/6/2013 12:24 AMID:148797
People who viewed this also viewed

Feedback

Did this KB document help you?



What can we do to improve this information (2000 or fewer characters)