How to avoid ERROR: The command request timed out. ?
Some times when trying to retrieve large amounts of data, my connector is showing message: ERROR: The command request timed out.
Is it possible to perform some configuration in order to increase the request time out ? any advice?
This might helpful to detect query previews: https://dev.scribesoft.com/en/sdk/html/p_scribe_core_connectorapi_query_...
Hi Nate, Thank you for the link. I was able to fix timeouts in the UI preview by using the suggested query.IsTestQuery property.
By the way, we are using JAVA Rest API that is already paginated. The connector was developed in C# with CDK. when IsTestQuery is TRUE then only first page is retrieved from the API (this way preview timeout was fixed) other wise the connector performs calls to the API for multiple pages. Now we are facing timeouts for On Demand / Scheduled executions with many pages of results involved.
Thanks for your time and help.
You may notice that the preview will timeout, where the query itself may be running without a timeout in the connector. The UI has a shorter timeout than many endpoints.
You may try paging the query to avoid the timeout. If it is http or another stream, there are ways to page these in smaller chunks to avoid timeouts and forcibly closed exceptions. Let me know the technology of your API and I will try to provide further detail.
Hi Nate, Thanks for your answer.
Just to clarify, the timeout message occurs when trying to preview information from the query block properties in the UI (On Demand and Scheduled executions seem to be working fine). This is happening for ON24 connectors which I have been developing.
So, is this the expected behavior for large amounts of data in the UI or do we need to perform some fix / configuration?
Thanks and regards,
Also, can you please confirm if from the connector's code perspective is it possible to detect if the incoming query comes from a preview (UI) instead of an onDemand/Scheduled execution, in order to reduce the number of records to be returned ?