Add Prometheus metrics support for DNS query monitoring #7937
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
![windsurf-bot[bot]](https://avatars.githubusercontent.com/in/1066231?s=60&v=4){kind=small}
Comment on lines
+192
to
+191
| // Register metrics endpoint without control prefix, similar to /dns-query | ||
| // Use empty method to bypass auth/gzip middleware like dns-query does | ||
| httpRegister("", "/metrics", web.metricsHandler.ServeHTTP) |
There was a problem hiding this comment.
While it makes sense to have the metrics endpoint unauthenticated like the /dns-query endpoint for Prometheus scraping, it would be good to document the security considerations of exposing metrics without authentication. Consider adding a comment explaining what information is exposed and any potential security implications.
- Add prometheus/client_golang dependency - Initialize Prometheus registry with Go and process collectors - Expose metrics endpoint at /metrics (without auth, similar to /dns-query) - Set up base infrastructure for future DNS filtering metrics
- Create metrics/dns.go with counters for total and blocked queries - Integrate metrics collection in stats unit.add() method - Register DNS metrics in web API Prometheus registry - Add comprehensive tests for metrics functionality Metrics exposed: - adguard_dns_queries_total: Total DNS queries processed - adguard_dns_queries_by_result_total: Queries by result type (filtered, not_filtered, etc.) - adguard_dns_queries_blocked_total: Legacy metric for blocked queries The metrics mirror the dashboard counters exactly, providing real-time DNS query statistics for monitoring and alerting.
- Add native exponential histogram for DNS response times - Track response times by result type (filtered, not_filtered, etc.) - Use factor 1.1 for automatic bucket generation - Integrate with existing stats processing flow
metalmatze
force-pushed
the
feature/prometheus-metrics
branch
from
2916d13 to
7058411
Compare
|
highly needed |
|
Yep. I'm running this custom build for months now. It works great. It would be fantastic to get a review! |
|
Do you have any updates on this? If there's interest in merging this, I will rebase. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
Closes #516
This PR implements native Prometheus metrics support for AdGuard Home, addressing the long-standing community request for built-in monitoring capabilities.
🚀 Features Added
/metricsendpoint (unauthenticated, similar to/dns-query)📊 Metrics Exposed
adguard_dns_queries_total{result}- Total DNS queries by processing resultadguard_dns_response_time_seconds{result}- DNS response time histograms🏗️ Implementation Details
internal/metrics/dns.gowith clean API for metric collectioninternal/stats/unit.go)🔧 Technical Approach
1.1for automatic bucket generation✅ Testing & Validation
Real Examples
I've deployed a binary with the changes in this PR on my NAS and put my home network's load on this instance.
Here we can see the
rate(adguard_dns_queries_total[5m])query. It shows the requests per second grouped by result type:Here we can see a
1000 * histogram_quantile(0.75, sum by(result) (rate(adguard_dns_response_time_seconds[5m])))query. It shows the milliseconds (hence1000 *) over time. 75% of requests by result type are faster than the given duration and 25% are slower.