bito-code-review[bot] commented on PR #43846: URL: https://github.com/apache/superset/pull/43846#issuecomment-5534516983
<!-- Bito Reply --> The suggestion is correct. The TTL for the async result cache begins when the worker writes the result to the cache, not when the task starts. Therefore, including the task execution time in the TTL requirement is unnecessary and overstates the required retention period. To resolve this, you should update the documentation to clarify that the TTL requirement is based on the time from result write to client retrieval, rather than the total round-trip time including execution. Would you like me to implement this fix in the documentation and check the other comments on this PR? **docs/admin_docs/configuration/cache.mdx** ``` This read-back is what makes the result-cache TTL matter for correctness, not just performance: if the effective TTL is shorter than the time from result write to client retrieval, the entry can be **evicted before the client reads it**, leaving the chart stuck re-running instead of loading. ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
