Skip to content

Automate browser based workflows with AI

chore(extraction-cache): promote cross-run miss log DEBUG → INFO (SKY-8992) (#5575)

pedrohsdb7f1533dcfed0
diff --git a/skyvern/webeye/actions/handler.py b/skyvern/webeye/actions/handler.py
index 590ccee06..320cd869e 100644
--- a/skyvern/webeye/actions/handler.py
+++ b/skyvern/webeye/actions/handler.py
@@ -4555,17 +4555,14 @@ async def extract_information_for_navigation_goal(
# does not track it yet.
return ScrapeResult(scraped_data=cross_run_value)
- # Cross-run miss log — DEBUG so it doesn't flood INFO at 0% rollout
- # (where the cloud override returns None for every call regardless of
- # wpid or Redis state). When the flag ramps past the initial measurement
- # checkpoint we can promote to INFO in the same commit that flips the
- # percentage. Paired with the cross-run hit log (still INFO) this
- # gives a computable hit rate at any rollout level.
- # TODO(SKY-8992): promote this log from DEBUG → INFO in the same commit
- # that flips the PostHog read flag past 0%, so the Datadog hit-rate
- # dashboard has both sides of the ratio without a log-level backfill.
+ # Cross-run miss log — INFO so the wpid-tier hit rate is computable
+ # from logs alone once the read flag starts ramping. Earlier drafts kept
+ # this at DEBUG specifically to avoid flooding INFO during the
+ # post-merge 0%-read window; promoted to INFO in SKY-8992 before the
+ # first read-flag flip so Datadog has both sides of the ratio without a
+ # log-level backfill.
if cache_key is not None and not is_retry_step and cross_run_value is None:
- LOG.debug(
+ LOG.info(
"extract_information cache miss (cross-run)",
task_id=task.task_id,
workflow_run_id=task.workflow_run_id,