Add 600s max histogram bucket for catalog_metadata_retrieval_latency metric #432
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…metric Configure histogram buckets to extend to 600 seconds for the metadata retrieval latency metric, enabling capture of long-running metadata operations. Changes: - Add maximum-expected-value.catalog_metadata_retrieval_latency=600s to application.properties - Add MetricsHistogramConfigurationTest to verify Spring Boot configuration
cbb330
force-pushed
the
cbb330-histogram-600s
branch
from
984464b to
235e7a7
Compare
teamurko
previously approved these changes
Jan 14, 2026
sumedhsakdeo
previously approved these changes
Jan 14, 2026
…eusMeterRegistry - Use @AutoConfigureMetrics to enable production metrics config in tests (Spring Boot disables metrics exporters by default in tests) - Validate MeterRegistry is PrometheusMeterRegistry, not SimpleMeterRegistry - Scrape /actuator/prometheus endpoint to verify le="600.0" bucket exists - Confirms maximum-expected-value.catalog_metadata_retrieval_latency=600s is applied
abhisheknath2011
previously approved these changes
Jan 15, 2026
Member
abhisheknath2011
left a comment
abhisheknath2011
left a comment
There was a problem hiding this comment.
Thanks for the PR. Have seen couple of instances of sudden latency spike earlier where some of the requests are actually stuck (at specific time range) and took significant amount of time while retrieving metadata. I have captured two instances of such high latency from HDFS in a doc.
Replace HTTP endpoint parsing with direct Timer.takeSnapshot() inspection: - Remove TestRestTemplate and /actuator/prometheus calls - Use HistogramSnapshot.histogramCounts() to verify bucket boundaries - Change @SpringBootTest to use default webEnvironment (faster startup) - Remove brittle Prometheus text format parsing
abhisheknath2011
approved these changes
Jan 15, 2026
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Critical latency metric is pinned to 30s when in reality it is as high as 5 minutes. Incorrect latency was interpreted in an investigation and led to wrong conclusions.
above: log based metric shows real latency
below: prometheus based metric shows p99 capped at 30s
Configure histogram buckets to extend to 600 seconds for the
catalog_metadata_retrieval_latencymetric, enabling accurate capture of long-running metadata operations.Changes
Added
maximum-expected-value.catalog_metadata_retrieval_latency=600sto application.properties and a new Spring Boot integration test to verify the configuration.Testing Done
Added
MetricsHistogramConfigurationTestwhich verifies:Timer.takeSnapshot().histogramCounts())Additional Information