My Tableau prep was updated to 2021.4 this morning. After the update, a flow of mine started encountering an error when refresh. The error message is "System error: AqlProcessor evaluation failed: No more data to read from socket". I was able to identify the error happened on an Aggregation step, I attached the Aggregate in the screenshot. Basically I added an output step right before and after the Aggregate 7 to test the location of failure, the output before Aggregate 7 worked but the output after Aggregate 7 failed.
However, when I put mouse on Aggregate 7, it shows the sample result. I'm not sure what does the error mean. It only happened after the tableau prep upgrade.
Please update to 2022.2 , they have resolved it
https://www.tableau.com/support/releases/prep/2022.2.1
When running a flow with a Join and Union step, it would sometimes result in the following error, "AqlProcessor evaluation failed: no such column "
1344295
Related
I am using the query below and it does not like it.
where captured_date (between to_number(dateadd(month,-3getdate()),'99999999') and to_number(dateadd(month,-1,getdate()),'99999999'))
Error Message: line 17:22: mismatched input 'between'. Expecting: <expression>
I am new to using Athena and I am not sure why it's not liking the queries I normally use.
I used the query below.
where captured_date (between to_number(dateadd(month,-3getdate()),'99999999') and to_number(dateadd(month,-1,getdate()),'99999999'))
I was expecting the last three months of data to be returned.
I’m running into a problem that might be self inflicted but I’m a noob and I’m still stumbling thru this dashboard I inherited.
We are using Grafana 8.1.5. I inherited a dashboard with a panel that is giving this error message:
“failed to convert long to wide series when converting from dataframe: can not convert to wide series, input has null time values” error.
I think I know the cause but I don’t know how to fix it. The data source is from a MySQL database. When we select “All” under the dropdown “Org”, we don’t get this error and we get 8 rows back and are expecting 8 rows.
When we select one org (in this example “Data Protection”) under the dropdown “Org”, we do get the error and we get 5 rows back. I think because 3 queries return 0 rows, we are getting this error.
Each pie slice has its own SQL query to get data. For the queries that return 0 rows, I tried selecting some of the “Standard options” to fix it but it doesn’t seem to work. I tried to play around with the Min, Max, and “No Value” settings and enter values larger than 0 but I still get this error. The initial values for Min and Max were "auto" and "No Value" was blank but the screenshot is when I left them as "1" while I was messing with them.
What do I need to do to make the message go away? Is the message benign and I shouldn’t worry about it since it looks like we’re still printing out the pie chart?
Write SQL, which returns record with zero value instead of empty result. In theory zero value, won't be visible in the pie chart and you will have still record for dataframe transformation, so that one won't be failing.
So I accidentally stumbled across the answer. What happened is the query returned data formatted as "Time Series" instead of "Table". Because some of the data didn't have values for the specified time, the time value was Null.
Since it was Null, the chart returned this error message. The fix was to select "Table" as the format. Then the chart won't care about the time value being Null and the missing data appears as 0 in the legend and we don't get the error message anymore.
I am using JasperSoft Reports v.6.2.1 and when running a report within the Studio preview the output comes after 2 seconds.
Running the same report (output xlsx) on the server takes > half a minute - though there is no data volume issue (crosstab, 500 lines, 17 columns in excel, "ignore pagination" = true).
I am using $P{LoggedInUsername} to filter data within the WHERE-part of a WITH-clause (based on the user's rights), run the report and realized, when using a fixed value (the user's id as a string) instead of the parameter in the query, the report execution speed is good.
Same against Oracle DB from SQL Developer - the query resultset with a user's id string is back in 2 sec.
Also the output of $P{LoggedInUsername} in a TextField produces a String.
Once switching back to the $P{LoggedInUsername}-parameter in the query, the report takes ages again or runs out of heap memory in the Studio/server.
What could be the issue?
Finally my problem was solved using the expression user_id = '$P!{LoggedInUsername}' instead of $P{LoggedInUsername} in the WHERE-part of my query.
I have executed a query in HIVE CLI that should generate around 11.000.000 rows, I know the result because I have executed the query in the MS SQL Server Management Studio too.
The problem is that in HIVE CLI the rows are showing on an on ( right know there are more than 12 hours since I started the execution ) and all I want to know is the time processing, which is showed only after showing the results.
So I have 2 questions :
How to skip showing rows results in HIVE command line ?
If I will execute the query in Beeswax, how do I see statistics like execution time , similar with SET STATISTICS TIME ON in T-SQL ?
You can check it using link given in log .But it wont give you total processing left.
Just force-closed my Tableau workbook full of metrics, and when I opened again, I received the error message:
'Error in Parameters for command 'get-quantative-color' bad value: fn
After this, I received error messages that all of my calculated fields and even some basic date fields 'do not exist,' and I am basically left with a blank workbook.
Happy to provide more detail if needed -- anyone have any suggestions? Have been working on the workbook for a while and would rather not have to recreate from scratch :)