Dataset refresh failure for Impala odbc data source

[#1]

Hello All,

We have installed the Cloudera ODBC driver of version 2.6.11 and configured ODBC driver. Making use of impala ODBC as a data source and custom query in Power BI desktop, developed a report and published it into the power bi service. But when i refreshed the dataset to get the updated data in the Power BI report. We are getting below error, can anyone help me to troubleshoot the refresh issue.

Error message 1

{"error":{"code":"DM_GWPipeline_Gateway_MashupDataAccessError","pbi.error":{"code":"DM_GWPipeline_Gateway_MashupDataAccessError","parameters":{},"details":[{"code":"DM_ErrorDetailNameCode_UnderlyingErrorCode","detail":{"type":1,"value":"-2147467259"}},{"code":"DM_ErrorDetailNameCode_UnderlyingErrorMessage","detail":{"type":1,"value":"<ccon>ODBC: ERROR [08S01] [Cloudera][ImpalaODBC] (450) Error when calling the Impala Thrift API ExecuteStatement: SSL_read: error code: 0</ccon>"}},{"code":"DM_ErrorDetailNameCode_UnderlyingHResult","detail":{"type":1,"value":"-2147467259"}},{"code":"Microsoft.Data.Mashup.ValueError.DataSourceKind","detail":{"type":1,"value":"Odbc"}},{"code":"Microsoft.Data.Mashup.ValueError.DataSourcePath","detail":{"type":1,"value":"<ccon>dsn=Impala PROD</ccon>"}},{"code":"Microsoft.Data.Mashup.ValueError.OdbcErrors","detail":{"type":1,"value":"#table({\"SQLState\", \"NativeError\", \"Message\"}, {})"}},{"code":"Microsoft.Data.Mashup.ValueError.Reason","detail":{"type":1,"value":"DataSource.Error"}}],"exceptionCulprit":1}}} Table: New Accounts and Web Changes.

Error message 2:

{"error":{"code":"DM_GWPipeline_Gateway_MashupDataAccessError","pbi.error":{"code":"DM_GWPipeline_Gateway_MashupDataAccessError","parameters":{},"details":[{"code":"DM_ErrorDetailNameCode_UnderlyingErrorCode","detail":{"type":1,"value":"-2147467259"}},{"code":"DM_ErrorDetailNameCode_UnderlyingErrorMessage","detail":{"type":1,"value":"<ccon>ODBC: ERROR [HY000] [Cloudera][ImpalaODBC] (110) Error while executing a query in Impala: [HY000] : Runtime Error: Admission for query exceeded timeout 900000ms in pool root.svcpbigateway. Queued reason: Not enough aggregate memory available in pool root.svcpbigateway with max mem resources 1024.00 GB (configured statically). Needed 232.00 GB but only 68.00 GB was available.\n</ccon>"}},{"code":"DM_ErrorDetailNameCode_UnderlyingHResult","detail":{"type":1,"value":"-2147467259"}},{"code":"Microsoft.Data.Mashup.ValueError.DataSourceKind","detail":{"type":1,"value":"Odbc"}},{"code":"Microsoft.Data.Mashup.ValueError.DataSourcePath","detail":{"type":1,"value":"<ccon>dsn=Impala PROD</ccon>"}},{"code":"Microsoft.Data.Mashup.ValueError.OdbcErrors","detail":{"type":1,"value":"#table({\"SQLState\", \"NativeError\", \"Message\"}, {})"}},{"code":"Microsoft.Data.Mashup.ValueError.Reason","detail":{"type":1,"value":"DataSource.Error"}}],"exceptionCulprit":1}}} Table: RSA Bill Payments.

Thanks in advance!

[#2]

@Ynr Welcome to the Cloudera Community!

To help you get the best possible solution, I have tagged our Impala experts @bbreak @ChethanYM  who may be able to assist you further.

Please keep us updated on your post, and we hope you find a satisfactory solution to your query.

[#3]

Hello @bbreak @ChethanYM,

Any suggestions to troubleshoot the refresh issue.

[#4]

Hi @Ynr 

Can you try the latest ODBC driver version? The error message 1 and 2 both are different and occurs for different scenarios, The error 1 shows some connection issue via driver to Impala, Do enable the trace level logging for driver to get more logs and also review the power bi logs since the issue is while refreshing in the power BI. Refer the similar article:

https://community.cloudera.com/t5/Support-Questions/SAS-Impala-connection-error-Impala-Thrift-API/m-...

The error 2 shows memory issue after you submitting the query, You may need to tune the impala admission control pool to avoid this error and have the enough memory in the pool as per your requirements.

https://docs.cloudera.com/runtime/7.2.0/impala-manage/topics/impala-admission.html

Regards,

Chethan YM

[#5]

Thanks for replying @ChethanYM .

For error 1 Do you know how to enable the trace level logging for driver to get more logs? If yes please do share the article.

For error 2 to tune the impala admission control pool, do i need to reach Impala admin team.

[#6]

@Ynr In the below doc refer the "Configuring Logging Options on Windows" section to understand the steps to enable trace/debug level logging.

https://docs.cloudera.com/documentation/other/connectors/impala-odbc/2-6-11/Cloudera-ODBC-Driver-for...

Regards,

Chethan YM

[#7]

@Ynr Has the reply helped resolve your issue? If so, please mark the appropriate reply as the solution, as it will make it easier for others to find the answer in the future. Thanks.

[#8]

Hi @DianaTorres , Reached out to Impala admin team to resolve the refresh issue in Power BI. Thanks for the support

[#9]

Hello @Ynr , please, share how you resolve this problem? Tks

[#10]

This is how to resolve this problem:

To resolve this issue, set this property to 0 and restart Impala:

CM > Impala > Configuration > Impala Daemon command-line safety valve:

-idle_client_poll_period_s=0

This is a startup flag, not a query option. Its default value is 30 seconds and that is why the session in the above excerpt was closed after 30 secs. By setting it to 0, Impala will not periodically check the client connection. The client connections will remain open until they are explicitly closed on the client applications' side.