Few agent telemetry improvements #29770
Conversation
* Finalize and categorize default metrics to emit split via 4 profiles * Eliminate Status-based telemetry * Add CLI `agent diagnose show-telemetry` command to dump emitted telemetry * Added host name Os info for agent telemetry metadata
iglendd
force-pushed
the
len.gamburg/agent-telemetry-improve
branch
from
d01681b to
dfe7d15
Compare
iglendd
added
the
qa/done
Regression DetectorRegression Detector ResultsRun ID: e3c6954e-2854-464a-8658-211d5f8e818b Metrics dashboard Target profiles Baseline: 596053e Performance changes are noted in the perf column of each table:
No significant changes in experiment optimization goalsConfidence level: 90.00% There were no significant changes in experiment optimization goals at this confidence level and effect size tolerance.
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | pycheck_lots_of_tags | % cpu utilization | +2.22 | [-0.24, +4.69] | 1 | Logs |
| ➖ | otel_to_otel_logs | ingress throughput | +0.32 | [-0.49, +1.14] | 1 | Logs |
| ➖ | basic_py_check | % cpu utilization | +0.16 | [-2.52, +2.84] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | +0.12 | [-0.61, +0.85] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.00 | [-0.11, +0.11] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | +0.00 | [-0.01, +0.01] | 1 | Logs |
| ➖ | idle_all_features | memory utilization | -0.11 | [-0.20, -0.03] | 1 | Logs |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | -0.17 | [-0.21, -0.12] | 1 | Logs |
| ➖ | idle | memory utilization | -0.26 | [-0.30, -0.21] | 1 | Logs |
| ➖ | file_tree | memory utilization | -0.87 | [-0.97, -0.76] | 1 | Logs |
Bounds Checks
| perf | experiment | bounds_check_name | replicates_passed |
|---|---|---|---|
| ✅ | idle | 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:
-
Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
-
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.
-
Its configuration does not mark it "erratic".
| metrics: | ||
| - name: oracle.activity_samples_count | ||
| - name: oracle.activity_latency | ||
| - name: oracle.statement_metrics | ||
| - name: oracle.statement_plan_errors |
There was a problem hiding this comment.
cc @sethsamuel (in upcoming in the 7.60, I have moved metrics to more distinct sections (aka profiles). This one is for DB
Test changes on VMUse this command from test-infra-definitions to manually test this PR changes on a VM: inv create-vm --pipeline-id=48466195 --os-family=ubuntuNote: This applies to commit 3501a68 |
What does this PR do?
agent diagnose show-telemetrycommand to dump emitted telemetryMotivation
Describe how to test/QA your changes
Possible Drawbacks / Trade-offs
n/a
Additional Notes
n/a