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.

Sign up for GitHub

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

Jump to bottom

Add preview_datadog_agent_sampling #6112

Merged
merged 36 commits into from
Dec 16, 2024
Merged

Add preview_datadog_agent_sampling #6112

merged 36 commits into from
Dec 16, 2024

Conversation

BrynCooke
Copy link
Contributor

@BrynCooke BrynCooke commented Oct 3, 2024 •
edited by abernix
Loading

Respect x-datadog-sampling-priority (PR #6017)

This PR consists of two fixes:

Datadog priority sampling resolution is not lost.

Previously a x-datadog-sampling-priority of -1 would be converted to 0 for downstream requests and 2 would be converted to 1.

The sampler option in the telemetry.exporters.tracing.common.sampler is not datadog aware.

To get accurate APM metrics all spans must be sent to the datadog agent with a psr or sampling.priority attribute set appropriately to record the sampling decision.

preview_datadog_agent_sampling option in the router.yaml enables this behavior and should be used when exporting to the datadog agent via OTLP or datadog native.

telemetry:
  exporters:
    tracing:
      common:
        # Only 10 percent of spans will be forwarded from the Datadog agent to Datadog. Experiment to find a value that is good for you!
        sampler: 0.1
        # Send all spans to the Datadog agent. 
        preview_datadog_agent_sampling: true

By using these options, you can decrease your Datadog bill as you will only be sending a percentage of spans from the Datadog agent to datadog.

Important

Users must enable preview_datadog_agent_sampling to get accurate APM metrics.

Important

The Router does not support in-agent ingestion control.
Configuring traces_per_second in the Datadog agent will NOT dynamically adjust the Router's sampling rate to meet the target rate.

Important

Sending all spans to the datadog agent may require that you tweak the batch_processor settings in your exporter config. This applies to both OTLP and the Datadog native exporter.

By @BrynCooke in #6017

Checklist

Complete the checklist (and note appropriate exceptions) before the PR is marked ready-for-review.

  • Changes are compatible1
  • Documentation2 completed
  • Performance impact assessed and acceptable
  • Tests added and passing3
    • Unit Tests
    • Integration Tests
    • Manual Tests

Exceptions

Note any exceptions here

Notes

Fixes #6355

Footnotes

  1. It may be appropriate to bring upcoming changes to the attention of other (impacted) groups. Please endeavour to do this before seeking PR approval. The mechanism for doing this will vary considerably, so use your judgement as to how and when to do this. ↩

  2. Configuration is an important part of many changes. Where applicable please try to document configuration examples. ↩

  3. Tick whichever testing boxes are applicable. If you are adding Manual Tests, please document the manual testing (extensively) in the Exceptions. ↩

@router-perf
Copy link

router-perf bot commented Oct 3, 2024

CI performance tests

  • connectors-const - Connectors stress test that runs with a constant number of users
  • const - Basic stress test that runs with a constant number of users
  • demand-control-instrumented - A copy of the step test, but with demand control monitoring and metrics enabled
  • demand-control-uninstrumented - A copy of the step test, but with demand control monitoring enabled
  • enhanced-signature - Enhanced signature enabled
  • events - Stress test for events with a lot of users and deduplication ENABLED
  • events_big_cap_high_rate - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity
  • events_big_cap_high_rate_callback - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity using callback mode
  • events_callback - Stress test for events with a lot of users and deduplication ENABLED in callback mode
  • events_without_dedup - Stress test for events with a lot of users and deduplication DISABLED
  • events_without_dedup_callback - Stress test for events with a lot of users and deduplication DISABLED using callback mode
  • extended-reference-mode - Extended reference mode enabled
  • large-request - Stress test with a 1 MB request payload
  • no-tracing - Basic stress test, no tracing
  • reload - Reload test over a long period of time at a constant rate of users
  • step-jemalloc-tuning - Clone of the basic stress test for jemalloc tuning
  • step-local-metrics - Field stats that are generated from the router rather than FTV1
  • step-with-prometheus - A copy of the step test with the Prometheus metrics exporter enabled
  • step - Basic stress test that steps up the number of users over time
  • xlarge-request - Stress test with 10 MB request payload
  • xxlarge-request - Stress test with 100 MB request payload

@BrynCooke BrynCooke changed the title Add experimental_datadog_agent_sampling Add preview_datadog_agent_sampling Oct 3, 2024
@BrynCooke BrynCooke force-pushed the bryn/datadog-agent-support branch from a5f6f0a to 928a9fa Compare October 3, 2024 19:01
@svc-apollo-docs
Copy link
Collaborator

svc-apollo-docs commented Oct 3, 2024 •
edited
Loading

✅ Docs Preview Ready

No new or changed pages found.

@BrynCooke BrynCooke force-pushed the bryn/datadog-agent-support branch from 928a9fa to b1040aa Compare October 3, 2024 19:08
Add experimental_datadog_agent_sampling
a1b638d 
This mode will change the behaviour of the router for tracing in the following ways:

* Spans are never dropped, instead they are converted to RecordOnly.
* Spans that are sent to otlp and datadog exporters will always look like they have been sampled.
* The `sampling.priority` attribute is populated on spans.
* `psr` is populated in trace state.
* `m` is populated in trace state.
@BrynCooke BrynCooke force-pushed the bryn/datadog-agent-support branch from b1040aa to a1b638d Compare October 3, 2024 19:12
@BrynCooke BrynCooke marked this pull request as ready for review October 3, 2024 19:18
@BrynCooke BrynCooke requested review from a team as code owners October 3, 2024 19:18
bnjjj
bnjjj requested changes Oct 4, 2024
View reviewed changes
.changesets/fix_bryn_datadog_upstream_sampling_decision_test.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_upstream_sampling_decision_test.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_upstream_sampling_decision_test.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_upstream_sampling_decision_test.md Outdated
Comment on lines 45 to 46
> [!IMPORTANT]
> Sending all spans to the datadog agent may require that you tweak the `batch_processor` settings in your exporter config. This applies to both OTLP and the Datadog native exporter.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should we also mention potential perf impacts ?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've added some documentation around setting the sampler:
https://deploy-preview-6112--apollo-router-docs.netlify.app/configuration/telemetry/exporters/tracing/datadog#sampler-configuration

apollo-router/src/plugins/telemetry/mod.rs Outdated Show resolved Hide resolved
@BrynCooke BrynCooke requested a review from shorgi October 4, 2024 09:58
@BrynCooke BrynCooke requested a review from bnjjj October 4, 2024 10:16
shorgi
shorgi reviewed Oct 16, 2024
View reviewed changes
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/overview.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/overview.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/overview.mdx Outdated Show resolved Hide resolved
docs/source/configuration/telemetry/exporters/tracing/datadog.mdx Outdated Show resolved Hide resolved
shorgi
shorgi reviewed Oct 16, 2024
View reviewed changes
.changesets/fix_bryn_datadog_agent_sampling.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_agent_sampling.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_agent_sampling.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_agent_sampling.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_agent_sampling.md Outdated Show resolved Hide resolved
.changesets/fix_bryn_datadog_upstream_sampling_decision_propagation.md Outdated Show resolved Hide resolved
BrynCooke and others added 6 commits November 19, 2024 09:39
@BrynCooke
Merge branch 'dev' into bryn/datadog-agent-support
8726737
@BrynCooke
Merge branch 'dev' into bryn/datadog-agent-support
8909929
@BrynCooke
Merge branch 'dev' into bryn/datadog-agent-support
1105d31
Improve datadog and otlp sampling tests
30ffb13 
We made some weird configuration previously to force particular behaviour which when looked at subsequently didn't make any sense.

In particular the otlp tests should have had otlp propagation with the datadog propagator also enabled.

Also added a chack for the subgraph service to ensure it is correctly sampled.
@BrynCooke
Merge branch 'dev' into bryn/datadog-agent-support
b8a8e20
Lint and test fixes
4490216
@BrynCooke BrynCooke mentioned this pull request Dec 11, 2024
Open
@BrynCooke BrynCooke force-pushed the bryn/datadog-agent-support branch from 4c83075 to a3f3e35 Compare December 13, 2024 14:15
@BrynCooke BrynCooke force-pushed the bryn/datadog-agent-support branch from a3f3e35 to 437f053 Compare December 13, 2024 14:18
@BrynCooke BrynCooke merged commit a12d988 into dev Dec 16, 2024
13 checks passed
@BrynCooke BrynCooke deleted the bryn/datadog-agent-support branch December 16, 2024 23:30
lrlna pushed a commit that referenced this pull request Dec 18, 2024
@BrynCooke @lrlna
Add preview_datadog_agent_sampling (#6112)
68fce74
lrlna pushed a commit that referenced this pull request Dec 18, 2024
@BrynCooke @lrlna
Add preview_datadog_agent_sampling (#6112)
9b62604
BrynCooke added a commit that referenced this pull request Dec 18, 2024
@BrynCooke
Add preview_datadog_agent_sampling (#6112)
4771765
BrynCooke added a commit that referenced this pull request Dec 18, 2024
@BrynCooke
Add preview_datadog_agent_sampling (#6112)
2347a6f
BrynCooke added a commit that referenced this pull request Dec 20, 2024
@BrynCooke
Add preview_datadog_agent_sampling (#6112)
76d0d02
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@shorgi shorgi shorgi left review comments

@bnjjj bnjjj bnjjj approved these changes

@IvanGoncharov IvanGoncharov Awaiting requested review from IvanGoncharov

Assignees

@BrynCooke BrynCooke

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Datadog spans explosion after v1.53.x
5 participants
@BrynCooke @svc-apollo-docs @bnjjj @shorgi @abernix