feat(byoo): gate collector config env vars

.github/workflows/bazel.yml

@@ -495,6 +495,9 @@ jobs:
           # the cacheless retry, so any future non-cache flag is applied to both.
           # CACHE holds only the cache-specific flags.
           COMMON=()
+          if [ "${{ matrix.subtree.id }}" = "byoo-otel-collector" ]; then
+            COMMON+=(--jobs=1)
+          fi
           CACHE=(--remote_cache=)
           if [ "${CACHE_READY:-0}" = "1" ]; then
             CACHE=(--remote_cache="$CACHE_ENDPOINT"
@@ -589,6 +592,9 @@ jobs:
           # They still compile in the build step; only their execution is
           # deferred to that lane (see the bazel-integration job).
           COMMON=(--flaky_test_attempts=3 --test_tag_filters=-requires-docker)
+          if [ "${{ matrix.subtree.id }}" = "byoo-otel-collector" ]; then
+            COMMON+=(--jobs=1)
+          fi
           CACHE=(--remote_cache=)
           if [ "${CACHE_READY:-0}" = "1" ]; then
             CACHE=(--remote_cache="$CACHE_ENDPOINT"

.github/workflows/byoo-otel-collector.yml

@@ -0,0 +1,152 @@
+# SPDX-FileCopyrightText: Copyright (c) NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+name: byoo-otel-collector
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+  merge_group:
+    types: [checks_requested]
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+defaults:
+  run:
+    shell: bash
+
+concurrency:
+  group: byoo-otel-collector-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  detect:
+    name: detect BYOO OTel collector changes
+    runs-on: ubuntu-latest
+    outputs:
+      changed: ${{ steps.detect.outputs.changed }}
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Detect changes
+        id: detect
+        env:
+          EVENT: ${{ github.event_name }}
+          REF: ${{ github.ref }}
+          BEFORE_SHA: ${{ github.event.before }}
+          HEAD_SHA: ${{ github.sha }}
+          BASE_REF: ${{ github.base_ref }}
+          BASE_SHA: ${{ github.event.pull_request.base.sha }}
+          MERGE_GROUP_BASE_SHA: ${{ github.event.merge_group.base_sha }}
+        run: |
+          set -euo pipefail
+
+          run_all=false
+          changed=""
+
+          case "$EVENT" in
+            workflow_dispatch)
+              run_all=true
+              ;;
+            pull_request)
+              if ! changed=$(git diff --name-only "${BASE_SHA}...HEAD" 2>/dev/null); then
+                echo "diff against pull request base failed; running BYOO OTel collector checks"
+                run_all=true
+              fi
+              ;;
+            push)
+              if [ -z "${BEFORE_SHA}" ] || [ "${BEFORE_SHA}" = "0000000000000000000000000000000000000000" ]; then
+                run_all=true
+              elif ! changed=$(git diff --name-only "${BEFORE_SHA}..${HEAD_SHA}" 2>/dev/null); then
+                echo "diff ${BEFORE_SHA}..${HEAD_SHA} failed; running BYOO OTel collector checks"
+                run_all=true
+              fi
+              ;;
+            merge_group)
+              if [ -n "${MERGE_GROUP_BASE_SHA}" ]; then
+                if ! changed=$(git diff --name-only "${MERGE_GROUP_BASE_SHA}...HEAD" 2>/dev/null); then
+                  echo "diff against merge group base failed; running BYOO OTel collector checks"
+                  run_all=true
+                fi
+              else
+                run_all=true
+              fi
+              ;;
+            *)
+              run_all=true
+              ;;
+          esac
+
+          if [ "$REF" = "refs/heads/main" ]; then
+            run_all=true
+          fi
+
+          if [ "$run_all" = "true" ]; then
+            echo "changed=true" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
+
+          if printf "%s\n" "$changed" | grep -qE '^(src/compute-plane-services/byoo-otel-collector/|tools/ci/check-byoo-otel-collector-version|tools/ci/test-check-byoo-otel-collector-version|\.github/workflows/byoo-otel-collector\.yml$)'; then
+            echo "changed=true" >> "$GITHUB_OUTPUT"
+          else
+            echo "changed=false" >> "$GITHUB_OUTPUT"
+          fi
+
+  version:
+    name: check VERSION update
+    needs: detect
+    if: needs.detect.outputs.changed == 'true'
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Run version gate tests
+        run: ./tools/ci/test-check-byoo-otel-collector-version
+
+      - name: Check BYOO OTel collector VERSION changed
+        env:
+          EVENT: ${{ github.event_name }}
+          BEFORE_SHA: ${{ github.event.before }}
+          HEAD_SHA: ${{ github.sha }}
+          BASE_REF: ${{ github.base_ref }}
+          BASE_SHA: ${{ github.event.pull_request.base.sha }}
+          MERGE_GROUP_BASE_SHA: ${{ github.event.merge_group.base_sha }}
+        run: ./tools/ci/check-byoo-otel-collector-version
+
+  validate:
+    name: validate configs (${{ matrix.mode }})
+    needs: detect
+    if: needs.detect.outputs.changed == 'true'
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        mode:
+          - vm-container
+          - vm-helm
+          - k8s-container
+          - k8s-helm
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-go@v5
+        with:
+          # The generated otelcol module says `go 1.25.0`, but Linux
+          # builds of OTel v0.157.0 need the patched toolchain line used
+          # by this subtree's Bazel module.
+          go-version-file: src/compute-plane-services/byoo-otel-collector/go.work
+
+      - name: Validate generated collector configs
+        working-directory: src/compute-plane-services/byoo-otel-collector
+        env:
+          GOCACHE: ${{ runner.temp }}/go-cache
+          GO_BUILD_P: "1"
+        run: ./scripts/validate-otelconfig.sh "${{ matrix.mode }}"

docs/ngc-managed/cluster-management/configuration.md

@@ -733,6 +733,28 @@ pod and merged with the generated config at runtime.
        logLevel: debug
 ```
 
+BYOO collector debug and log chunking example:
+
+```yaml
+ agentConfig:
+   mergeConfig: |
+     agent:
+       byooLogChunking:
+         enabled: true
+         maxPayloadBytes: 262144
+       byooDebugMode:
+         enabled: true
+       byooOtelCollector:
+         exporterHelper:
+           timeout: 30s
+           sendingQueue:
+             batch:
+               flushTimeout: 200ms
+               sizer: bytes
+               minSize: 1000000
+               maxSize: 1000000
+```
+
 BYOO metric subset example:
 
 ```yaml

docs/user/cluster-management/configuration.md

@@ -854,6 +854,28 @@ isolated test clusters that run the LLM request router tunnel without TLS. For
 the full LLM addon setup, see
 [LLM Function Enablement](../llm-function-enablement.md).
 
+BYOO collector debug and log chunking example:
+
+```yaml
+agentConfig:
+  mergeConfig: |
+    agent:
+      byooLogChunking:
+        enabled: true
+        maxPayloadBytes: 262144
+      byooDebugMode:
+        enabled: true
+      byooOtelCollector:
+        exporterHelper:
+          timeout: 30s
+          sendingQueue:
+            batch:
+              flushTimeout: 200ms
+              sizer: bytes
+              minSize: 1000000
+              maxSize: 1000000
+```
+
 BYOO metric subset example:
 
 ```yaml

src/compute-plane-services/byoo-otel-collector/AGENTS.md

@@ -34,17 +34,17 @@ otelconfig validation. Do not add a subtree `.gitlab-ci.yml`.
 Use the script instead of editing version strings by hand:
 
 ```bash
-./scripts/update-collector-version.sh v0.153.0 v1.59.0
+./scripts/update-collector-version.sh v0.157.0 v1.63.0
 # If service release tags are already ahead of the collector patch:
-./scripts/update-collector-version.sh v0.153.0 v1.59.0 0.153.2
+./scripts/update-collector-version.sh v0.157.0 v1.63.0 0.153.2
 ```
 
 The script updates version references in `otel-collector-build.yaml`,
 `AGENTS.md`, `README.md`, `Makefile`, `Dockerfile`,
 `Dockerfile.nvcf-otel-collector`, `scripts/regenerate-otelcol.sh`, and
 `VERSION`. It also updates `.gitlab-ci.yml` when that file exists. Run it from
 the BYOO collector root. You can pass versions with or without the `v` prefix
-(for example, `v0.153.0` or `0.153.0`). Pass the optional `v1.x.y` provider
+(for example, `v0.157.0` or `0.157.0`). Pass the optional `v1.x.y` provider
 version when the stable collector modules need a matching release, and pass the
 optional app release override when existing service tags require the next BYOO
 patch version. The app release major/minor must match the collector

src/compute-plane-services/byoo-otel-collector/Dockerfile.nvcf-otel-collector

@@ -40,7 +40,7 @@ RUN cd otelcol && \
 FROM nvcr.io/nvidia/distroless/go:v4.0.4 AS nvcf-otel-collector
 
 LABEL description="NVCF OpenTelemetry Collector - Custom build with curated components for NVCF services"
-LABEL version="0.153.0"
+LABEL version="0.157.0"
 
 COPY --from=builder-otelcol /app/output/otelcol-contrib /otelcol-contrib
 

src/compute-plane-services/byoo-otel-collector/MODULE.bazel

@@ -83,6 +83,7 @@ use_repo(
     "com_github_kelseyhightower_envconfig",
     "com_github_open_telemetry_opentelemetry_collector_contrib_exporter_azuremonitorexporter",
     "com_github_open_telemetry_opentelemetry_collector_contrib_exporter_datadogexporter",
+    "com_github_open_telemetry_opentelemetry_collector_contrib_exporter_prometheusexporter",
     "com_github_open_telemetry_opentelemetry_collector_contrib_exporter_prometheusremotewriteexporter",
     "com_github_open_telemetry_opentelemetry_collector_contrib_exporter_splunkhecexporter",
     "com_github_open_telemetry_opentelemetry_collector_contrib_extension_basicauthextension",

src/compute-plane-services/byoo-otel-collector/Makefile

@@ -1,7 +1,7 @@
 .PHONY: test docker-build docker-run update-examples update-config-template validate-config
 
 DOCKER_TAG := $(shell whoami)-dev
-OTEL_BUILDER_VERSION ?= v0.153.0
+OTEL_BUILDER_VERSION ?= v0.157.0
 
 test:
 	find . -name go.mod -execdir go test -v ./... \;