Date Posted:
Product: TIBCO Spotfire®
Product: TIBCO Spotfire®
Problem:
Information Link becomes unavailable after restarting SAS/SHARE Service: "Connection reset by peer: socket write error."
Solution:
Symptoms:
Below error is returned while trying to open the datasource :
Error: InformationModelException at Spotfire.Dxp.Data:
Error retrieving metadata: Sharexxx..xxx: Sharexxx.xxx.ex.txt: Connection reset by peer: socket write error (HRESULT: 80131500)Cause:
SAS/SHARE does not support connection pooling. This issue is generally noticed when connection pooling is enabled on the SAS/SHARE Data Source in Spotfire.
Resolution 1:
============
Turn off the connection pooling for SAS/SHARE Data Source to resolve the issue.
1). Login to Spotfire Analyst and Go to Tools -> Information Designer
2). Edit the data source for SAS/SHARE.
3). Set the Min and Max value for "No of connections" parameter to "0" and save.
Here is an extract from the Spotfire Analyst help manual:
Note: For SAS/SHARE and ODBC data sources, both Min and Max number of connections should be set to 0 in order to disable connection pooling. The same holds for other data sources that do not support pooled connection since there is no valid ping command.
Note: For SAS/SHARE and ODBC data sources, both Min and Max number of connections should be set to 0 in order to disable connection pooling. The same holds for other data sources that do not support pooled connection since there is no valid ping command.
==========
Update your SAS/Share Data source template with the below ping command i.e. update line <ping-command/> to <ping-command>SELECT 1 FROM DUAL</ping-command>
<jdbc-type-settings>
<type-name>sas/share</type-name>
<driver>com.sas.net.sharenet.ShareNetDriver</driver>
<connection-url-pattern>jdbc:sharenet://<host>:<port></connection-url-pattern>
<ping-command>SELECT 1 FROM DUAL</ping-command>
<supports-catalogs>false</supports-catalogs>
<supports-schemas>true</supports-schemas>
<supports-procedures>false</supports-procedures>
<table-types>DATA, VIEW</table-types>
<column-name-pattern>$$name$$</column-name-pattern>
<table-name-pattern>$$name$$</table-name-pattern>
<schema-name-pattern>$$name$$</schema-name-pattern>
<column-alias-pattern>$$name$$</column-alias-pattern>
</jdbc-type-settings>
https://www.labkey.org/home/Developer/issues/issues-details.view?issueId=8828