perf(sidecar)!: Batch ack sending & consumption#1835
Conversation
Clippy Allow Annotation ReportComparing clippy allow annotations between branches:
Summary by Rule
Annotation Counts by File
Annotation Stats by Crate
About This ReportThis report tracks Clippy allow annotations for specific rules, showing how they've changed in this PR. Decreasing the number of these annotations generally improves code quality. |
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #1835 +/- ##
==========================================
+ Coverage 71.55% 71.60% +0.04%
==========================================
Files 426 426
Lines 67155 67287 +132
==========================================
+ Hits 48053 48181 +128
- Misses 19102 19106 +4
🚀 New features to boost your workflow:
|
|
✅ Tests 🎉 All green!❄️ No new flaky tests detected 🎯 Code Coverage (details) 🔗 Commit SHA: bca1af9 | Docs | Datadog PR Page | Was this helpful? React with 👍/👎 or give us feedback! |
Artifact Size Benchmark Reportaarch64-alpine-linux-musl
aarch64-unknown-linux-gnu
libdatadog-x64-windows
libdatadog-x86-windows
x86_64-alpine-linux-musl
x86_64-unknown-linux-gnu
|
Leiyks
left a comment
There was a problem hiding this comment.
few things worth nothing but overall looks good 👍
| #[cfg(target_os = "linux")] | ||
| { | ||
| __pending_acks += 1; | ||
| if #force_flush || __pending_acks >= 20 { |
There was a problem hiding this comment.
Would be cleaner to add a constant like ACK_BUFFER_SIZE and reuse it in blocking.rs for consistency
| }; | ||
| let mut telemetry_guard = telemetry_mutex.lock_or_panic(); | ||
| let Some(telemetry) = telemetry_guard.as_mut() else { | ||
| return; // extremely rare: stopped again between the two get_or_create calls |
There was a problem hiding this comment.
We should add a log here even if it is rare just in case
The newly introduced acks always require an additional syscall to consume them. Using recvmmsg() to batch it. Also only consume them when we actually need to check for limits. Also avoid resending Start each time we access the TelemetryCachedClient. (Pure overhead) Signed-off-by: Bob Weinand <bob.weinand@datadoghq.com>
Signed-off-by: Bob Weinand <bob.weinand@datadoghq.com>
Signed-off-by: Bob Weinand <bob.weinand@datadoghq.com>
Signed-off-by: Bob Weinand <bob.weinand@datadoghq.com>
Signed-off-by: Bob Weinand <bob.weinand@datadoghq.com>
Signed-off-by: Bob Weinand <bob.weinand@datadoghq.com>
|
/merge |
|
View all feedbacks in Devflow UI.
This pull request is not mergeable according to GitHub. Common reasons include pending required checks, missing approvals, or merge conflicts — but it could also be blocked by other repository rules or settings.
The expected merge time in
|
# Release proposal for libdd-telemetry and its dependencies This PR contains version bumps based on public API changes and commits since last release. ## libdd-capabilities **Next version:** `2.0.0` **Semver bump:** `major` **Tag:** `libdd-capabilities-v2.0.0` ### Commits - feat(capablities)!: sleep & spawn capabilities (#1873) ## libdd-common **Next version:** `4.1.0` **Semver bump:** `minor` **Tag:** `libdd-common-v4.1.0` ### Commits - refactor(sampling): move the sampling logic from dd-trace-rs [APMSP-2946] (#1927) - feat!: added regex-lite feature (#1939) - fix(libdd-common): crashes caused by `getenv` while retrieving AAS env vars (#1930) ## libdd-capabilities-impl **Next version:** `2.0.0` **Semver bump:** `major` **Tag:** `libdd-capabilities-impl-v2.0.0` ### Commits - feat(capablities)!: sleep & spawn capabilities (#1873) ## libdd-shared-runtime **Next version:** `1.0.0` **Semver bump:** `major` **Tag:** `libdd-shared-runtime-v1.0.0` **Warning:** this is an initial release. Please verify that the version and commits included are correct. ## libdd-telemetry **Next version:** `5.0.0` **Semver bump:** `major` **Tag:** `libdd-telemetry-v5.0.0` ###⚠️ major bump forced due to: - `libdd-common`: ^3.0.2 → ^4.0.0 ### Commits - fix(libdd-telemetry): restore previous Cargo.toml version (#1993) - feat(capablities)!: sleep & spawn capabilities (#1873) - fix(telemetry): avoid trigger loop in telemetry worker (#1950) - feat(telemetry)!: include dependencies and integrations in app-extended-heartbeat (#1962) - fix(crypto): gate libdd-common TLS features in remaining internal crates + add CI guard (#1943) - fix(telemetry): schedule ExtendedHeartbeat on worker start (#1910) - fix(telemetry): skip sending empty payloads (#1894) - feat(sidecar): wire telemetry_extended_heartbeat_interval through SessionConfig (#1882) - chore(telemetry): add session id support (#1817) - ci(libdd-shared-runtime): downgrade version so publish workflow succeeds (#1870) - feat(runtime)!: add shared runtime (#1602) - perf(sidecar)!: Batch ack sending & consumption (#1835) - fix(telemetry): wire up DD_TELEMETRY_EXTENDED_HEARTBEAT_INTERVAL to scheduler (#1824) - feat(capabilities)!: trait architecture http (#1555) - chore(telemetry): use weaker mem ordering for SEQ_ID (#1749) [APMSP-2946]: https://datadoghq.atlassian.net/browse/APMSP-2946?atlOrigin=eyJpIjoiNWRkNTljNzYxNjVmNDY3MDlhMDU5Y2ZhYzA5YTRkZjUiLCJwIjoiZ2l0aHViLWNvbS1KU1cifQ --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: iunanua <18325288+iunanua@users.noreply.github.com>
The newly introduced acks always require an additional syscall to consume them.
Using sendmmsg()/recvmmsg() to batch it. Also only consume them when we actually need to check for limits.
Also avoid resending Start each time we access the TelemetryCachedClient. (Pure overhead)