[CONTP-60] Improved telemetry on cluster check configs dangling #32508
Conversation
gabedos
added
the
qa/done
gabedos
added
team/container-platform
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: 5bd602d ❌ Experiments with missing or malformed dataThis is a critical error. No usable optimization goal data was produced by the listed experiments. This may be a result of misconfiguration. Ping #single-machine-performance and we can help out.
Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | quality_gate_logs | % cpu utilization | +4.84 | [+1.46, +8.22] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | +0.17 | [-0.53, +0.86] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | +0.04 | [-0.60, +0.68] | 1 | Logs |
| ➖ | otel_to_otel_logs | ingress throughput | +0.04 | [-0.61, +0.68] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.01 | [-0.10, +0.12] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | +0.01 | [-0.81, +0.84] | 1 | Logs |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | +0.01 | [-0.76, +0.78] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | +0.00 | [-0.01, +0.01] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http1 | egress throughput | -0.01 | [-0.90, +0.89] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http2 | egress throughput | -0.03 | [-0.97, +0.92] | 1 | Logs |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | -0.03 | [-0.72, +0.66] | 1 | Logs |
| ➖ | file_tree | memory utilization | -0.13 | [-0.26, -0.00] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.17 | [-0.97, +0.63] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | -0.18 | [-0.64, +0.28] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | -0.25 | [-0.33, -0.16] | 1 | Logs bounds checks dashboard |
| ➖ | quality_gate_idle | memory utilization | -0.67 | [-0.70, -0.63] | 1 | Logs bounds checks dashboard |
Bounds Checks: ❌ Failed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ❌ | file_to_blackhole_0ms_latency_http1 | lost_bytes | 9/10 | |
| ✅ | file_to_blackhole_0ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency_linear_load | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | memory_usage | 10/10 | |
| ✅ | quality_gate_idle | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_logs | lost_bytes | 10/10 | |
| ✅ | quality_gate_logs | memory_usage | 10/10 |
Explanation
Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%
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
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".
CI Pass/Fail Decision
✅ Passed. All Quality Gates passed.
- quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check lost_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
gabedos
force-pushed
the
gabedos/extend-config-dangling
branch
from
66a1af8 to
e7b7ccb
Compare
Uncompressed package size comparisonComparison with ancestor Diff per package
Decision |
Test changes on VMUse this command from test-infra-definitions to manually test this PR changes on a VM: inv aws.create-vm --pipeline-id=51849592 --os-family=ubuntuNote: This applies to commit ddcdf78 |
gabedos
force-pushed
the
gabedos/extend-config-dangling
branch
from
20a1bb3 to
eea7b33
Compare
gabedos
force-pushed
the
gabedos/extend-config-dangling
branch
from
b2fb193 to
3a1c7af
Compare
|
/merge |
Devflow running:
|
What does this PR do?
Adds a new telemetry metric to keep track of cluster check configs that have gone an extended period of time without being scheduled.
Motivation
We would like to monitor the state of the cluster-check scheduling in order to be sure we detect issues before applications get paged for missing metrics.
Describe how you validated your changes
Existing unit tests cover dispatching logic. Update to a unit test verifies the unscheduledCheck flag behavior.
Possible Drawbacks / Trade-offs
N/A
Additional Notes
Previously, on each rescheduling attempt, all the dangling configs were deleted from the dangling store initially. If they failed to be rescheduled, then they would be readded the dangling config store.
This PR changes this behavior because we want to track how long the config exists in the dangling state. Hence, we only remove the config from the dangling store once it's confirmed to be scheduled.