Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[smp]: move smp junit upload to new job #30198

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

goxberry
Copy link
Contributor

What does this PR do?

While working with a colleague on another task, I noticed that fetching secrets took a surprisingly long time. Moving secrets fetching and our JUnit XML upload operations to a separate step helps clean up the signal we get from Pipeline Visibility regarding Regression Detector job execution times.

Motivation

More meaningful statistics from SMP's Test Visibility monitoring.

Describe how to test/QA your changes

SMP tests should continue to be uploaded to Test Visibility.

Possible Drawbacks / Trade-offs

We use Pipeline Visibility to get a quick aggregate view of SMP's health in Agent CI -- particularly to understand if Agent CI jobs are terminating due to timeouts. If secrets fetching or uploading JUnit XML take nontrivial amounts of time (say, more than a dozen seconds each), then we mistakenly attribute GitLab CI timeouts to under-provisioning of SMP runners in our backend, potentially causing us to overallocate capacity. Moving the JUnit XML upload step to a separate job should fix that attribution problem.

Additional Notes

n/a

While working with a colleague on another task, I noticed that
fetching secrets took a surprisingly long time. Moving secrets
fetching and our JUnit XML upload operations to a separate step helps
clean up the signal we get from Pipeline Visibility regarding
Regression Detector job execution times.

Signed-off-by: Geoffrey M. Oxberry <[email protected]>
@goxberry goxberry added changelog/no-changelog team/single-machine-performance Single Machine Performance qa/no-code-change No code change in Agent code requiring validation labels Oct 16, 2024
@goxberry goxberry requested a review from a team as a code owner October 16, 2024 23:28
@agent-platform-auto-pr
Copy link
Contributor

Gitlab CI Configuration Changes

Modified Jobs

single-machine-performance-regression_detector
  single-machine-performance-regression_detector:
    allow_failure: true
    artifacts:
      expire_in: 1 weeks
      paths:
      - submission_metadata
      - ${CI_COMMIT_SHA}-baseline_sha
      - outputs/report.md
      - outputs/regression_signal.json
      - outputs/bounds_check_signal.json
      - outputs/junit.xml
      when: always
    image: 486234852809.dkr.ecr.us-east-1.amazonaws.com/ci/datadog-agent-buildimages/docker_x64$DATADOG_AGENT_BUILDIMAGES_SUFFIX:$DATADOG_AGENT_BUILDIMAGES
    needs:
    - artifacts: false
      job: single_machine_performance-amd64-a7
    rules:
    - if: $CI_COMMIT_BRANCH == "main"
      when: never
    - if: $CI_COMMIT_BRANCH =~ /^[0-9]+\.[0-9]+\.x$/
      when: never
    - if: $CI_COMMIT_BRANCH =~ /^mq-working-branch-/
      when: never
    - when: on_success
    script:
    - mkdir outputs
    - touch outputs/report.md
    - git fetch origin
    - SMP_BASE_BRANCH=$(inv release.get-release-json-value base_branch)
    - echo "Looking for merge base for branch ${SMP_BASE_BRANCH}"
    - SMP_MERGE_BASE=$(git merge-base ${CI_COMMIT_SHA} origin/${SMP_BASE_BRANCH})
    - echo "Merge base is ${SMP_MERGE_BASE}"
    - AWS_NAMED_PROFILE="single-machine-performance"
    - SMP_ACCOUNT_ID=$($CI_PROJECT_DIR/tools/ci/fetch_secret.sh $SMP_ACCOUNT_ID) ||
      exit $?
    - SMP_ECR_URL=${SMP_ACCOUNT_ID}.dkr.ecr.us-west-2.amazonaws.com
    - SMP_AGENT_TEAM_ID=$($CI_PROJECT_DIR/tools/ci/fetch_secret.sh $SMP_AGENT_TEAM_ID)
      || exit $?
    - SMP_API=$($CI_PROJECT_DIR/tools/ci/fetch_secret.sh $SMP_API) || exit $?
    - SMP_BOT_ID=$($CI_PROJECT_DIR/tools/ci/fetch_secret.sh $SMP_BOT_ACCESS_KEY_ID)
      || exit $?
    - SMP_BOT_KEY=$($CI_PROJECT_DIR/tools/ci/fetch_secret.sh $SMP_BOT_ACCESS_KEY) ||
      exit $?
    - aws configure set aws_access_key_id "$SMP_BOT_ID" --profile ${AWS_NAMED_PROFILE}
    - aws configure set aws_secret_access_key "$SMP_BOT_KEY" --profile ${AWS_NAMED_PROFILE}
    - aws configure set region us-west-2 --profile ${AWS_NAMED_PROFILE}
    - aws --profile single-machine-performance s3 cp s3://smp-cli-releases/v${SMP_VERSION}/x86_64-unknown-linux-gnu/smp
      smp
    - chmod +x smp
    - BASELINE_SHA="${SMP_MERGE_BASE}"
    - echo "Computing baseline..."
    - echo "Checking if commit ${BASELINE_SHA} is recent enough..."
    - FOUR_DAYS_BEFORE_NOW=$(date --date="-4 days +1 hour" "+%s")
    - BASELINE_COMMIT_TIME=$(git -c log.showSignature=false show --no-patch --format=%ct
      ${BASELINE_SHA})
    - "if [[ ${BASELINE_COMMIT_TIME} -le ${FOUR_DAYS_BEFORE_NOW} ]]\nthen\n    echo\
      \ \"ERROR: Merge-base of this branch is too old for SMP. Please update your branch\
      \ by merging an up-to-date main branch into your branch or by rebasing it on an\
      \ up-to-date main branch.\"\n    exit 1\nfi\n"
    - echo "Commit ${BASELINE_SHA} is recent enough"
    - echo "Checking if image exists for commit ${BASELINE_SHA}..."
    - "while [[ ! $(aws ecr describe-images --profile single-machine-performance --registry-id\
      \ \"${SMP_ACCOUNT_ID}\" --repository-name \"${SMP_AGENT_TEAM_ID}-agent\" --image-ids\
      \ imageTag=\"${BASELINE_SHA}-7-amd64\") ]]\ndo\n    echo \"No image exists for\
      \ ${BASELINE_SHA} - checking predecessor of ${BASELINE_SHA} next\"\n    BASELINE_SHA=$(git\
      \ rev-parse ${BASELINE_SHA}^)\n    echo \"Checking if commit ${BASELINE_SHA} is\
      \ recent enough...\"\n    BASELINE_COMMIT_TIME=$(git -c log.showSignature=false\
      \ show --no-patch --format=%ct ${BASELINE_SHA})\n    if [[ ${BASELINE_COMMIT_TIME}\
      \ -le ${FOUR_DAYS_BEFORE_NOW} ]]\n    then\n        echo \"ERROR: Merge-base of\
      \ this branch is too old for SMP. Please update your branch by merging an up-to-date\
      \ main branch into your branch or by rebasing it on an up-to-date main branch.\"\
      \n        exit 1\n    fi\n    echo \"Commit ${BASELINE_SHA} is recent enough\"\
      \n    echo \"Checking if image exists for commit ${BASELINE_SHA}...\"\ndone\n"
    - echo "Image exists for commit ${BASELINE_SHA}"
    - echo "Baseline SHA is ${BASELINE_SHA}"
    - echo -n "${BASELINE_SHA}" > "${CI_COMMIT_SHA}-baseline_sha"
    - aws s3 cp --profile single-machine-performance --only-show-errors "${CI_COMMIT_SHA}-baseline_sha"
      "s3://${SMP_AGENT_TEAM_ID}-smp-artifacts/information/"
    - BASELINE_IMAGE=${SMP_ECR_URL}/${SMP_AGENT_TEAM_ID}-agent:${BASELINE_SHA}-7-amd64
    - echo "${BASELINE_SHA} | ${BASELINE_IMAGE}"
    - COMPARISON_IMAGE=${SMP_ECR_URL}/${SMP_AGENT_TEAM_ID}-agent:${CI_COMMIT_SHA}-7-amd64
    - echo "${CI_COMMIT_SHA} | ${COMPARISON_IMAGE}"
    - RUST_LOG="info,aws_config::profile::credentials=error"
    - RUST_LOG_DEBUG="debug,aws_config::profile::credentials=error"
    - RUST_LOG="${RUST_LOG}" ./smp --team-id ${SMP_AGENT_TEAM_ID} --api-base ${SMP_API}
      --aws-named-profile ${AWS_NAMED_PROFILE} job submit --baseline-image ${BASELINE_IMAGE}
      --comparison-image ${COMPARISON_IMAGE} --baseline-sha ${BASELINE_SHA} --comparison-sha
      ${CI_COMMIT_SHA} --target-config-dir test/regression/ --submission-metadata submission_metadata
    - RUST_LOG="${RUST_LOG}" ./smp --team-id ${SMP_AGENT_TEAM_ID} --api-base ${SMP_API}
      --aws-named-profile ${AWS_NAMED_PROFILE} job status --wait --wait-delay-seconds
      60 --submission-metadata submission_metadata
    - RUST_LOG="${RUST_LOG}" ./smp --team-id ${SMP_AGENT_TEAM_ID} --api-base ${SMP_API}
      --aws-named-profile ${AWS_NAMED_PROFILE} job sync --submission-metadata submission_metadata
      --output-path outputs
    - cat outputs/report.md | sed "s/^\$/$(echo -ne '\uFEFF\u00A0\u200B')/g"
-   - DATADOG_API_KEY="$("$CI_PROJECT_DIR"/tools/ci/fetch_secret.sh "$AGENT_API_KEY_ORG2"
-     token)" || exit $?; export DATADOG_API_KEY
-   - datadog-ci junit upload --service datadog-agent outputs/junit.xml
    - RUST_LOG="${RUST_LOG}" ./smp --team-id ${SMP_AGENT_TEAM_ID} --api-base ${SMP_API}
      --aws-named-profile ${AWS_NAMED_PROFILE} job result --submission-metadata submission_metadata
    stage: functional_test
    tags:
    - runner:docker
    timeout: 1h10m
    variables:
      SMP_VERSION: 0.16.0

Added Jobs

single-machine-performance-regression_detector-junit-upload
single-machine-performance-regression_detector-junit-upload:
  allow_failure: true
  image: 486234852809.dkr.ecr.us-east-1.amazonaws.com/ci/datadog-agent-buildimages/docker_x64$DATADOG_AGENT_BUILDIMAGES_SUFFIX:$DATADOG_AGENT_BUILDIMAGES
  needs:
  - job: single-machine-performance-regression_detector
  rules:
  - if: $CI_COMMIT_BRANCH == "main"
    when: never
  - if: $CI_COMMIT_BRANCH =~ /^[0-9]+\.[0-9]+\.x$/
    when: never
  - if: $CI_COMMIT_BRANCH =~ /^mq-working-branch-/
    when: never
  - when: on_success
  script:
  - DATADOG_API_KEY="$("$CI_PROJECT_DIR"/tools/ci/fetch_secret.sh "$AGENT_API_KEY_ORG2"
    token)" || exit $?; export DATADOG_API_KEY
  - datadog-ci junit upload --service datadog-agent outputs/junit.xml
  stage: functional_test
  tags:
  - runner:docker

Changes Summary

Removed Modified Added Renamed
0 1 1 0

ℹ️ Diff available in the job log.

@agent-platform-auto-pr
Copy link
Contributor

[Fast Unit Tests Report]

On pipeline 46779226 (CI Visibility). The following jobs did not run any unit tests:

Jobs:
  • tests_deb-arm64-py3
  • tests_deb-x64-py3
  • tests_flavor_dogstatsd_deb-x64
  • tests_flavor_heroku_deb-x64
  • tests_flavor_iot_deb-x64
  • tests_rpm-arm64-py3
  • tests_rpm-x64-py3
  • tests_windows-x64

If you modified Go files and expected unit tests to run in these jobs, please double check the job logs. If you think tests should have been executed reach out to #agent-devx-help

Copy link

Regression Detector

Regression Detector Results

Run ID: c856f8ff-2adc-4d73-b07a-a20c25a37db0 Metrics dashboard Target profiles

Baseline: b0ca9e6
Comparison: 1fa34db

Performance changes are noted in the perf column of each table:

  • ✅ = significantly better comparison variant performance
  • ❌ = significantly worse comparison variant performance
  • ➖ = no significant change in performance

No significant changes in experiment optimization goals

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

There were no significant changes in experiment optimization goals at this confidence level and effect size tolerance.

Fine details of change detection per experiment

perf experiment goal Δ mean % Δ mean % CI trials links
basic_py_check % cpu utilization +1.02 [-1.66, +3.69] 1 Logs
idle_all_features memory utilization +0.74 [+0.65, +0.83] 1 Logs bounds checks dashboard
tcp_syslog_to_blackhole ingress throughput +0.50 [+0.45, +0.55] 1 Logs
idle memory utilization +0.48 [+0.43, +0.52] 1 Logs bounds checks dashboard
otel_to_otel_logs ingress throughput +0.35 [-0.45, +1.16] 1 Logs
file_to_blackhole_300ms_latency egress throughput +0.06 [-0.12, +0.24] 1 Logs
uds_dogstatsd_to_api ingress throughput +0.02 [-0.08, +0.12] 1 Logs
tcp_dd_logs_filter_exclude ingress throughput -0.00 [-0.01, +0.01] 1 Logs
file_to_blackhole_500ms_latency egress throughput -0.00 [-0.25, +0.25] 1 Logs
file_to_blackhole_0ms_latency egress throughput -0.01 [-0.34, +0.33] 1 Logs
file_to_blackhole_100ms_latency egress throughput -0.01 [-0.23, +0.21] 1 Logs
file_to_blackhole_1000ms_latency egress throughput -0.05 [-0.52, +0.42] 1 Logs
file_tree memory utilization -0.71 [-0.83, -0.59] 1 Logs
uds_dogstatsd_to_api_cpu % cpu utilization -0.79 [-1.50, -0.08] 1 Logs
pycheck_lots_of_tags % cpu utilization -3.05 [-5.42, -0.67] 1 Logs

Bounds Checks

perf experiment bounds_check_name replicates_passed
file_to_blackhole_0ms_latency memory_usage 10/10
file_to_blackhole_1000ms_latency memory_usage 10/10
file_to_blackhole_100ms_latency memory_usage 10/10
file_to_blackhole_300ms_latency memory_usage 10/10
file_to_blackhole_500ms_latency memory_usage 10/10
idle memory_usage 10/10
idle_all_features memory_usage 10/10

Explanation

A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".

For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:

  1. Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.

  2. Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.

  3. Its configuration does not mark it "erratic".

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
changelog/no-changelog qa/no-code-change No code change in Agent code requiring validation team/single-machine-performance Single Machine Performance
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant