moomindani commented on PR #14440: URL: https://github.com/apache/iceberg/pull/14440#issuecomment-5610529101
Not stale — @rdblue's 5 June review is addressed in full, and this has waited three months for a second look. 56/56 green, mergeable, one commit, core-only. Correcting my own approval of 9 August: I said the code uses `cache.expiration.expire-after-write-interval-ms` and asked for the thread with @findepi to be updated to match. That was wrong, quoted from an earlier revision. The only property is `cache.expiration-after-write-interval-ms`, and `CatalogProperties` and the docs table already agree on it. Nothing to fix there. @pvary in December you wrote that this goes in the right direction and suggested a community sync. The real-world evidence missing then is now in [issuecomment-4621071011](https://github.com/apache/iceberg/pull/14440#issuecomment-4621071011): with `expireAfterAccess=3s` a Spark catalog served stale data on 841 consecutive reads across a 300-second window and never refreshed, because every read resets the access timer, and engine-managed SQL leaves nowhere to call `table.refresh()`. A review, or a slot on a sync agenda, would settle it. The bot closes this in a week, and it already auto-closed once in March. @rdblue nothing from your list is outstanding. -- 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]
