Chris White
Chris White
@wangxing-git would you mind testing against [this branch](https://github.com/PrefectHQ/prefect/pull/17722) and letting me know if it resolves the issue for you? Also, if anyone else could test against this branch it would...
Hi @majo-aqfer - excellent write-up of your issue, thank you. A few quick comments: - `result_storage_key` configures persistence for _all_ persisted results, which currently includes failed payloads; we _do_ want...
@Hafeez-Aqfer ah sorry, the reason this is still re-executing is that your cache key is actually not static -- both `task_run.id` and `task_run.name` are generated _per task function invocation_. To...
@Hafeez-Aqfer you should _not_ use `result_storage_key` for your use case right now, as it currently stores failure states and instead should use a `cache_key_fn` (copying the same code from above...
@majo-aqfer if you use a static cache key function (meaning it should depend on values that do not change across task invocations) you will get your expected result; here is...
@Hafeez-Aqfer none of your shared code examples have used a static cache key which is why I keep emphasizing it.
@Hafeez-Aqfer that sample code does _not_ have a static cache key; it is using the `task_run.name` field which changes per invocation as I said in my previous comment.
@Hafeez-Aqfer we are talking past each other. If you upgrade to 3.1.2. the original problem will go away regardless of the cache key or result storage key that you use.
This appears to have been closed upstream
lol holy semantic versioning