I am not a Cognos expert, but I am trying to get Cognos to query Impala via JDBC correctly.
Setup:
Cognos 11.1.7-2106251648 (FP3)
Impala (cdh) 3.2.0
I tried various Impala JDBC drivers from cloudera (JDBC4, 41, 42) versions (2.6.17, 2.6.18, 2.6.23)
And I also tried Hive JDBC driver in Cognos to connect to Impala
I have the following issue:
When I run the report, Cognos deploys two queries into impala (I can see it in Impala query logs).
When the first query is done and Cognos fetches the first batch of rows, Cognos stops the connection and runs the query again.
Hive driver - cancels the first query after N seconds (when the first fetch is done), the cancel is successful and then the Cognos runs query second time and fetches all data
Impala driver - cancels the first query after N seconds (when the first fetch is done), the cancel is NOT successful (the query runs in Impala, but Cognos disconnects) and then the query runs a second time and Cognos fetches all data. The issue is that the query still runs in Impala and gets dropped after approx. 5minutes.
I think there is some issue, that Cognos is sending the query twice.
Edit1:
I am using dynamic query mode in Cognos.
I enabled full logging for a few minutes and I found following
The cognos is doing something called
14:01:41 borrowConnection in(JDBC)
14:01:49 borrowConnection out(JDBC)
14:01:49 borrowConnection in(JDBC)
times 14:01:41 and 14:01:49 are the exact times when first and second (same) query were executed in Impala
Related
I have a Blueprism object which opens an Oracle database connection and runs a SQL script. This object has been working correctly for the past 12 months. I am developing a new process and the object isn't bringing back any results. If I run the SQL select against the database it returns results. When Blueprism runs it doesn't bring back any results.
I am manually connecting as the same user account.
I cannot provide SQL but it is a standard Select with where clauses.
Stack Overflow is not allowing me to upload a photo of my object.
Issue was linked to how Blueprism queries dates. I needed to amend SQL query to use this syntax
between to_date('01/Feb/2017','dd/mm/yyyy') and to_date('31/Dec/2017','dd/mm/yyyy')
I'm running a complicated query against a Redshift cluster in which there are 4 tables used with some of them have billions of rows, and I get the following error:
failed to make a valid plan
If I limit the data, the query will run successfully.
-The Original query was an Oracle query which I've made some modifications on it, and data loaded in the tables in Redshift was also exported from Oracle.
-The query has a lot of JOINs and sub queries.
With those being said, going through the sub-queries one at a time, one of them didn't return any results, and that was the cause of this error in my case.
Fixing that particular sub-query and the main query accordingly, it worked successfully.
I am working in Microsoft Visual Studio to create my RDL's. I have a sub report which refuses to run when we put it up on SSRS. This report runs fine inside of Visual studio in preview mode even when pointed at a copy of the prod database (still takes 30 min to run but it completes). The report only returns 1 row with counts of a large amount of data for a summary.
The full error text is:
An error has occurred during report processing. (rsProcessingAborted)
o Cannot read the next data row for the dataset DataSet1. (rsErrorReadingNextDataRow)
A severe error occurred on the current command. The results, if any, should be discarded. Operation cancelled by user.
This report used to work but the query was not correctly pulling the information. I had to change the query and expand what it was pulling from the database. Is there any way this could be caused by not enough memory being given to SSRS? We are using SSRS 2008 r2.
Turns out my problem was solved by putting in the recommended indexes provided by SQL Query Analyzer. After the indexes were created the query ran in ~4 min no problems.
My colleague created a project in Talend to read data from Oracle database.
I used his project and so I have his Job context with connection parameters to Oracle DB and Talend successfully connects on that computer.
I've created a trivial job which is composed of two components: tOracleInput which should be reading data and tLogRow which should be redirecting output to Talend's terminal.
The problem is that when I start the job - data is not outputted to terminal and instead of row amount outputted per second I see Starting ... status.
Would it be connection issues, inappropriate java version on my computer or something else?
Starting... status means that the query is being executed. Usually it takes a few seconds to execute a simple query against the database. This is because of the Oracle database behavior that it starts to return the data without completing a full table scan. To use this feature you can use joins and filters, but not group by / order by.
On the other hand if you're using a view or executing a complex query, or just simply use DISTINCT it could happen that the query execution takes a few minutes. This is because the oracle database generates the ResultSet on the database side before returning the records.
I am using queries.xml file which contains 5 sql statements.
I would like to know if the 5 sql queries are executed at the same time or one sql query at a time when running the forge process?
The queries are executed one at a time as the file is processed. Also, the file is processed top down.