feat(spans): initial MongoDB description scrubbing support

CHANGELOG.md

@@ -23,6 +23,7 @@
 - Collect SDK information in profile chunks. ([#3882](https://github.com/getsentry/relay/pull/3882))
 - Introduce `trim = "disabled"` type attribute to prevent trimming of spans. ([#3877](https://github.com/getsentry/relay/pull/3877))
 - Make the tcp listen backlog configurable and raise the default to 1024. ([#3899](https://github.com/getsentry/relay/pull/3899))
+- Add experimental support for MongoDB query normalization. ([#3912](https://github.com/getsentry/relay/pull/3912))
 - Extract `user.geo.country_code` into span indexed. ([#3911](https://github.com/getsentry/relay/pull/3911))
 - Add `span.system` tag to span metrics ([#3913](https://github.com/getsentry/relay/pull/3913))
 - Switch glob implementations from `regex` to `regex-lite`. ([#3926](https://github.com/getsentry/relay/pull/3926))

relay-cabi/src/processing.rs

@@ -12,6 +12,7 @@ use std::sync::OnceLock;
 use chrono::{DateTime, Utc};
 use relay_cardinality::CardinalityLimit;
 use relay_dynamic_config::{normalize_json, GlobalConfig, ProjectConfig};
+use relay_event_normalization::span::description::ScrubMongoDescription;
 use relay_event_normalization::{
     normalize_event, validate_event, BreakdownsConfig, ClientHints, EventValidationConfig,
     GeoIpLookup, NormalizationConfig, RawUserAgentInfo,
@@ -273,6 +274,7 @@ pub unsafe extern "C" fn relay_store_normalizer_normalize_event(
         normalize_spans: config.normalize_spans,
         replay_id: config.replay_id,
         span_allowed_hosts: &[], // only supported in relay
+        scrub_mongo_description: ScrubMongoDescription::Disabled, // only supported in relay
     };
     normalize_event(&mut event, &normalization_config);
 

relay-dynamic-config/src/defaults.rs

@@ -8,13 +8,7 @@ use crate::metrics::MetricSpec;
 use crate::{Feature, GroupKey, MetricExtractionConfig, ProjectConfig, Tag, TagMapping};
 
 /// A list of `span.op` patterns that indicate databases that should be skipped.
-const DISABLED_DATABASES: &[&str] = &[
-    "*clickhouse*",
-    "*compile*",
-    "*mongodb*",
-    "*redis*",
-    "db.orm",
-];
+const DISABLED_DATABASES: &[&str] = &["*clickhouse*", "*compile*", "*redis*", "db.orm"];
 
 /// A list of `span.op` patterns we want to enable for mobile.
 const MOBILE_OPS: &[&str] = &[
@@ -121,9 +115,9 @@ pub fn hardcoded_span_metrics() -> Vec<(GroupKey, Vec<MetricSpec>, Vec<TagMappin
     let is_ai = RuleCondition::glob("span.op", "ai.*");
 
     let is_db = RuleCondition::eq("span.sentry_tags.category", "db")
-        & !(RuleCondition::eq("span.system", "mongodb")
-            | RuleCondition::glob("span.op", DISABLED_DATABASES)
-            | RuleCondition::glob("span.description", MONGODB_QUERIES));
+        & !RuleCondition::glob("span.op", DISABLED_DATABASES)
+        & (RuleCondition::eq("span.system", "mongodb")
+            | !RuleCondition::glob("span.description", MONGODB_QUERIES));
     let is_resource = RuleCondition::glob("span.op", RESOURCE_SPAN_OPS);
 
     let is_cache = RuleCondition::glob("span.op", CACHE_SPAN_OPS);

relay-dynamic-config/src/feature.rs

@@ -99,6 +99,13 @@ pub enum Feature {
     #[serde(rename = "organizations:indexed-spans-extraction")]
     ExtractSpansFromEvent,
 
+    /// Enables description scrubbing for MongoDB spans (and consequently, their presence in the
+    /// Queries module inside Sentry).
+    ///
+    /// Serialized as `organizations:performance-queries-mongodb-extraction`.
+    #[serde(rename = "organizations:performance-queries-mongodb-extraction")]
+    ScrubMongoDBDescriptions,
+
     /// Deprecated, still forwarded for older downstream Relays.
     #[doc(hidden)]
     #[serde(rename = "organizations:transaction-name-mark-scrubbed-as-sanitized")]

relay-event-normalization/src/event.rs

@@ -30,6 +30,7 @@ use uuid::Uuid;
 
 use crate::normalize::request;
 use crate::span::ai::normalize_ai_measurements;
+use crate::span::description::ScrubMongoDescription;
 use crate::span::tag_extraction::extract_span_tags_from_event;
 use crate::utils::{self, get_event_user_tag, MAX_DURATION_MOBILE_MS};
 use crate::{
@@ -158,6 +159,9 @@ pub struct NormalizationConfig<'a> {
 
     /// Controls list of hosts to be excluded from scrubbing
     pub span_allowed_hosts: &'a [Host],
+
+    /// Controls whether or not MongoDB span descriptions will be scrubbed.
+    pub scrub_mongo_description: ScrubMongoDescription,
 }
 
 impl<'a> Default for NormalizationConfig<'a> {
@@ -190,6 +194,7 @@ impl<'a> Default for NormalizationConfig<'a> {
             normalize_spans: true,
             replay_id: Default::default(),
             span_allowed_hosts: Default::default(),
+            scrub_mongo_description: ScrubMongoDescription::Disabled,
         }
     }
 }
@@ -332,6 +337,7 @@ fn normalize(event: &mut Event, meta: &mut Meta, config: &NormalizationConfig) {
             event,
             config.max_tag_value_length,
             config.span_allowed_hosts,
+            config.scrub_mongo_description.clone(),
         );
     }