Hi GoodData team I just wanted to reach out regarding a `jav GoodData #gooddata-cloud

Hi GoodData team, I just wanted to reach out rega...

Mira Jill Yumul

10/27/2025, 2:29 AM

Hi GoodData team, I just wanted to reach out regarding a

java.sql.SQLNonTransientConnectionException

issue occurring in one of our visualizations. We would like to have a Barrister Overview table that lists the financial metrics (total invoices, outstanding, receipts, etc.) per barrister. In this scenario, it seems that if we add the 'Matter Count' metric, the error appears. Could you please help check and advise whether this is due to the size of the data, or complexity of the query? Thanks in advance 🙏 Link: https://barbooks-au.cloud.gooddata.com/analyze/#/1cb8f28d23d94c71b970f255178d0c4b/a3704fed-cfaa-4321-8876-5d9468443880/edit/reload

Julius Kos

10/27/2025, 10:41 AM

Hello Mira, it seems to work for me: Could you please double-check again?

👀 1

Mira Jill Yumul

10/27/2025, 11:48 PM

Hi Julius, Apologies for the late response. You're right, it is working on my end as well. Thank you for checking 🙂

Mira Jill Yumul

10/27/2025, 11:54 PM

I would also like to note, that when I change the order of the 'Matter Count' to be at the very first (from the initial workspace), the visualization takes a longer time to load, and it results in an error (traceId: 94f4d19ead83ff055cb758cacd6328d6) Could you help confirm why this is so?

Julius Kos

10/28/2025, 8:38 AM

Yes, this might be expected as the order of measures in the “Metrics (in columns)” panel really does matter for query planning. The first metric becomes the “driver” for the SQL the engine builds and everything else has to join to that subquery, which can lead to timeouts under heavier workloads. Actually, looking on your traceID - this is exactly what is happening here:

Copy code

"ts":"2025-10-27 23:52:12.999", "DATA_ERROR","detail":"Query timeout occurred"}, "traceId":"94f4d19ead83ff055cb758cacd6328d6"

📝 1

Julius Kos

10/28/2025, 8:41 AM

Btw, I'm not sure if you are aware but while being in some specific visualization - you can put

/debug

in the end of the URL (instead of

/edit

) and it will download a debug file where you can find the specific SQL query which is triggered against your data source. That way, you can analyze why some SQL on the backend could be promlematic in the context of the setup of the visualization.

🙌 1

📝 1

Mira Jill Yumul

10/29/2025, 1:11 AM

Thanks for the helpful response, Julius!

8 Views

Open in Slack

Previous Next