The GitHub Actions job "Java CI with Maven" on stormcrawler.git/status-updater-cache-counters has succeeded. Run started by GitHub user jnioche (triggered by jnioche).
Head commit for run: b70afdcff302d12f9394aa5e9ea6675880488d44 / Julien Nioche <[email protected]> AbstractStatusUpdaterBolt: report cache hits/misses with a counter, not reset-on-read gauges The "cache.hits" and "cache.misses" gauges returned the counts and reset them to 0 on every read. That works with the V1 metrics API, where a single consumer reads each time bucket, but not with V2: Dropwizard gauges must be side-effect free, and each V2 reporter (Console, CSV, JMX, Graphite, a Prometheus endpoint scraped by several servers or curled by hand) took the counts away from the others. They are now a scoped counter "cache" with scopes "hits" and "misses", registered through CrawlerMetrics.registerCounter. The metric names are unchanged: V2 registers cache.hits / cache.misses, and V1 consumers flatten the MultiCountMetric map to the same names, still per time bucket. "cache.size" stays a gauge. Under V2, cache.hits and cache.misses are now cumulative counters. Dashboards that used them as per-interval values should switch to rate() / increase(). Co-Authored-By: Claude Opus 5.5 <[email protected]> Report URL: https://github.com/apache/stormcrawler/actions/runs/38085662021 With regards, GitHub Actions via GitBox
