[e2e] create a session output directory per TestSuite session #32536
Conversation
Remove dependency to runner
Remove dependency from client to e2e
pducolin
added
changelog/no-changelog
qa/no-code-change
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=51843632 --os-family=ubuntuNote: This applies to commit df2f7be |
|
[Fast Unit Tests Report] On pipeline 51843632 (CI Visibility). The following jobs did not run any unit tests: Jobs:
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 |
Uncompressed package size comparisonComparison with ancestor Diff per package
Decision |
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: 237b1f2 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | quality_gate_logs | % cpu utilization | +2.54 | [-0.72, +5.80] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | +1.11 | [+0.42, +1.80] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http1 | egress throughput | +0.19 | [-0.70, +1.09] | 1 | Logs |
| ➖ | file_tree | memory utilization | +0.09 | [-0.04, +0.22] | 1 | Logs |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | +0.05 | [-0.72, +0.82] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | +0.02 | [-0.63, +0.66] | 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 |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | -0.01 | [-0.72, +0.71] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http2 | egress throughput | -0.01 | [-0.95, +0.93] | 1 | Logs |
| ➖ | quality_gate_idle | memory utilization | -0.08 | [-0.11, -0.04] | 1 | Logs bounds checks dashboard |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | -0.09 | [-0.96, +0.78] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | -0.31 | [-0.39, -0.23] | 1 | Logs bounds checks dashboard |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | -0.46 | [-0.52, -0.40] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.56 | [-1.35, +0.23] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | -0.62 | [-1.08, -0.16] | 1 | Logs |
Bounds Checks: ❌ Failed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ❌ | file_to_blackhole_0ms_latency | lost_bytes | 8/10 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | lost_bytes | 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 memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check lost_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
pducolin
force-pushed
the
pducolin/ADXT-354-part3-refactor-suite-utils-to-common
branch
2 times, most recently
from
90d1dc3 to
2200dd8
Compare
| parentDir := filepath.Dir(fullPath) | ||
| finalDir := filepath.Base(fullPath) |
There was a problem hiding this comment.
❓ stupid question: isn't parentDir the same as rootOutputDir and finalDir the same as subdirectory in this context? I guess my question is: why do we need to build fullPath?
There was a problem hiding this comment.
No question is stupid!
It depends on the passed subdirectory. As the current code uses the go test name, that contains path separators, and we don't sanitize them, we end up potentially with fullPath that is composed of
- the output directory
e2e-output - the suite session directory
<suite_name>_<suite.startTime>
The suite name can contain/if it is defined in a subfolder, for example this one from installer suite
I need to first build the parent directory, meaning all the path except the final directory, that will be a temporary one. This is to keep compatibility with current directory structure used in Windows tests: I would remove the session and temporary path for CI runs, as there is one session per job. This was implemented to ease local runs.
Remove not needed CreateTestSession dir function, centralise session output dir creation within a suite
pducolin
force-pushed
the
pducolin/ADXT-354-part3-refactor-suite-utils-to-common
branch
from
2200dd8 to
df2f7be
Compare
What does this PR do?
Refactor the session output directory management, centralising it inside the base e2e suite. Root output directory creation is delegated to the runner.profile, that is aware of running on the CI or locally.
This PR also reduces dependencies to
e2epackage out of itself, moving common code tocommonpackageMotivation
Refactoring required to implement
Diagnosableinterface for Host environmentDescribe how you validated your changes
Possible Drawbacks / Trade-offs
Additional Notes
Example of output directories for
pre_e2e_testsjobAs we use
suite.T().Name(), suites running from a singleTestXXXwill all end up with the same output dir prefixTestXXX, we might want to add the suite name to ease identifying each suite's session, might change it in a follow up