Merge branch 'master' into zerobus-defer-descriptor

Author: thomasqueirozb

.github/workflows/publish.yml

@@ -273,7 +273,7 @@ jobs:
 
   deb-verify:
     name: Verify DEB Packages
-    runs-on: ubuntu-24.04
+    runs-on: ${{ matrix.runner }}
     timeout-minutes: 5
     needs:
       - generate-publish-metadata
@@ -284,6 +284,9 @@ jobs:
       DD_PKG_VERSION: "latest"
     strategy:
       matrix:
+        runner:
+          - ubuntu-24.04
+          - ubuntu-24.04-arm
         container:
           - ubuntu:20.04
           - ubuntu:22.04
@@ -304,25 +307,29 @@ jobs:
           make
       - name: Install dd-pkg for linting
         run: |
-          curl -sSL "https://dd-package-tools.s3.amazonaws.com/dd-pkg/${DD_PKG_VERSION}/dd-pkg_Linux_x86_64.tar.gz" | tar -xz -C /usr/local/bin dd-pkg
+          ARCH=$(uname -m)
+          case "$ARCH" in aarch64) ARCH_DL=arm64 ;; *) ARCH_DL="$ARCH" ;; esac
+          curl -sSL "https://dd-package-tools.s3.amazonaws.com/dd-pkg/${DD_PKG_VERSION}/dd-pkg_Linux_${ARCH_DL}.tar.gz" | tar -xz -C /usr/local/bin dd-pkg
       - name: Fix Git safe directories issue when in containers (actions/checkout#760)
         run: git config --global --add safe.directory /__w/vector/vector
       - name: Checkout Vector
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
         with:
           ref: ${{ inputs.git_ref }}
-      - name: Download staged package artifacts (x86_64-unknown-linux-gnu)
+      - name: Set artifact name
+        run: echo "ARTIFACT=$(uname -m)-unknown-linux-gnu" >> "$GITHUB_ENV"
+      - name: Download staged package artifacts
         uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
         with:
-          name: vector-${{ env.VECTOR_VERSION }}-x86_64-unknown-linux-gnu
+          name: vector-${{ env.VECTOR_VERSION }}-${{ env.ARTIFACT }}
           path: target/artifacts
       - name: Verify install of DEB package.
         run: |
-          ./scripts/verify-install.sh target/artifacts/vector_${{ env.VECTOR_VERSION }}-1_amd64.deb
+          ./scripts/verify-install.sh "target/artifacts/vector_${{ env.VECTOR_VERSION }}-1_$(dpkg --print-architecture).deb"
 
   rpm-verify:
     name: Verify RPM Packages
-    runs-on: ubuntu-24.04
+    runs-on: ${{ matrix.runner }}
     timeout-minutes: 10
     needs:
       - generate-publish-metadata
@@ -333,6 +340,9 @@ jobs:
       DD_PKG_VERSION: "latest"
     strategy:
       matrix:
+        runner:
+          - ubuntu-24.04
+          - ubuntu-24.04-arm
         container:
           # Lowest supported glibc tier (EL8 family, glibc 2.28). Catches glibc-floor regressions.
           - "almalinux:8@sha256:ca871b110064fe81bff0250ed5cc56fbb0a9840715f09811c33bbf9d72cef56c"
@@ -358,21 +368,25 @@ jobs:
           fi
       - name: Install dd-pkg for linting
         run: |
-          curl -sSL "https://dd-package-tools.s3.amazonaws.com/dd-pkg/${DD_PKG_VERSION}/dd-pkg_Linux_x86_64.tar.gz" | tar -xz -C /usr/local/bin dd-pkg
+          ARCH=$(uname -m)
+          case "$ARCH" in aarch64) ARCH_DL=arm64 ;; *) ARCH_DL="$ARCH" ;; esac
+          curl -sSL "https://dd-package-tools.s3.amazonaws.com/dd-pkg/${DD_PKG_VERSION}/dd-pkg_Linux_${ARCH_DL}.tar.gz" | tar -xz -C /usr/local/bin dd-pkg
       - name: Fix Git safe directories issue when in containers (actions/checkout#760)
         run: git config --global --add safe.directory /__w/vector/vector
       - name: Checkout Vector
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
         with:
           ref: ${{ inputs.git_ref }}
-      - name: Download staged package artifacts (x86_64-unknown-linux-gnu)
+      - name: Set artifact name
+        run: echo "ARTIFACT=$(uname -m)-unknown-linux-gnu" >> "$GITHUB_ENV"
+      - name: Download staged package artifacts
         uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
         with:
-          name: vector-${{ env.VECTOR_VERSION }}-x86_64-unknown-linux-gnu
+          name: vector-${{ env.VECTOR_VERSION }}-${{ env.ARTIFACT }}
           path: target/artifacts
       - name: Verify install of RPM package.
         run: |
-          ./scripts/verify-install.sh target/artifacts/vector-${{ env.VECTOR_VERSION }}-1.x86_64.rpm
+          ./scripts/verify-install.sh "target/artifacts/vector-${{ env.VECTOR_VERSION }}-1.$(uname -m).rpm"
 
   macos-verify:
     name: Verify macOS Package

.github/workflows/semantic.yml

@@ -127,6 +127,7 @@ jobs:
             aws service
             azure service
             confluent service
+            databricks_zerobus sink
             datadog service
             elastic service
             gcp service

changelog.d/tag_cardinality_limit_improve_memory_usage.enhancement.md

@@ -0,0 +1,3 @@
+Reduce the memory usage of the `tag_cardinality_limit` transform when running in `exact` mode by allocating less unused memory on initialization.
+
+authors: ArunPiduguDD

src/transforms/tag_cardinality_limit/mod.rs

@@ -213,7 +213,7 @@ impl TagCardinalityLimit {
         let metric_accepted_tags = self.accepted_tags.entry(metric_key_owned).or_default();
         let tag_value_set = metric_accepted_tags
             .entry_ref(key)
-            .or_insert_with(|| AcceptedTagValueSet::new(config.value_limit, &config.mode));
+            .or_insert_with(|| AcceptedTagValueSet::new(&config.mode));
 
         if tag_value_set.contains(value) {
             // Tag value has already been accepted, nothing more to do.
@@ -300,7 +300,7 @@ impl TagCardinalityLimit {
         let metric_accepted_tags = self.accepted_tags.entry(metric_key_owned).or_default();
         metric_accepted_tags
             .entry_ref(key)
-            .or_insert_with(|| AcceptedTagValueSet::new(config.value_limit, &config.mode))
+            .or_insert_with(|| AcceptedTagValueSet::new(&config.mode))
             .insert(value.clone());
         false
     }

src/transforms/tag_cardinality_limit/tag_value_set.rs

@@ -59,9 +59,9 @@ impl fmt::Debug for TagValueSetStorage {
 }
 
 impl AcceptedTagValueSet {
-    pub fn new(value_limit: usize, mode: &Mode) -> Self {
+    pub fn new(mode: &Mode) -> Self {
         let storage = match &mode {
-            Mode::Exact => TagValueSetStorage::Set(HashSet::with_capacity(value_limit)),
+            Mode::Exact => TagValueSetStorage::Set(HashSet::new()),
             Mode::Probabilistic(config) => {
                 TagValueSetStorage::Bloom(BloomFilterStorage::new(config.cache_size_per_key))
             }
@@ -100,7 +100,7 @@ mod tests {
 
     #[test]
     fn test_accepted_tag_value_set_exact() {
-        let mut accepted_tag_value_set = AcceptedTagValueSet::new(2, &Mode::Exact);
+        let mut accepted_tag_value_set = AcceptedTagValueSet::new(&Mode::Exact);
 
         assert!(!accepted_tag_value_set.contains(&TagValueSet::from(["value1".to_string()])));
         assert_eq!(accepted_tag_value_set.len(), 0);
@@ -116,7 +116,7 @@ mod tests {
 
     #[test]
     fn test_accepted_tag_value_set_probabilistic() {
-        let mut accepted_tag_value_set = AcceptedTagValueSet::new(2, &Mode::Exact);
+        let mut accepted_tag_value_set = AcceptedTagValueSet::new(&Mode::Exact);
 
         assert!(!accepted_tag_value_set.contains(&TagValueSet::from(["value1".to_string()])));
         assert_eq!(accepted_tag_value_set.len(), 0);