[CWS] Refactor process resolver test #32306
Conversation
…ts to reflect new approach
github-actions
bot
added
component/system-probe
long review
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: 13597f4 Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | tcp_syslog_to_blackhole | ingress throughput | +0.59 | [+0.53, +0.65] | 1 | Logs |
| ➖ | file_tree | memory utilization | +0.40 | [+0.27, +0.53] | 1 | Logs |
| ➖ | quality_gate_idle | memory utilization | +0.21 | [+0.18, +0.24] | 1 | Logs bounds checks dashboard |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | +0.17 | [-0.29, +0.64] | 1 | Logs |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | +0.08 | [-0.70, +0.86] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http2 | egress throughput | +0.03 | [-0.87, +0.92] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | +0.01 | [-0.63, +0.65] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.01 | [-0.11, +0.12] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | +0.00 | [-0.08, +0.09] | 1 | Logs bounds checks dashboard |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.00 | [-0.01, +0.01] | 1 | Logs |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | -0.04 | [-0.74, +0.66] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | -0.05 | [-0.95, +0.85] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http1 | egress throughput | -0.07 | [-0.86, +0.71] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.13 | [-0.91, +0.65] | 1 | Logs |
| ➖ | otel_to_otel_logs | ingress throughput | -0.81 | [-1.48, -0.14] | 1 | Logs |
| ➖ | quality_gate_logs | % cpu utilization | -1.24 | [-4.49, +2.00] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | -1.92 | [-2.60, -1.24] | 1 | Logs |
Bounds Checks: ❌ Failed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ❌ | quality_gate_idle_all_features | memory_usage | 9/10 | bounds checks dashboard |
| ✅ | file_to_blackhole_0ms_latency | lost_bytes | 10/10 | |
| ✅ | 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_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
❌ Failed. Some Quality Gates were violated.
- quality_gate_idle_all_features, bounds check memory_usage: 9/10 replicas passed. Failed 1 which is > 0. Gate FAILED.
- 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.
mftoure
added
the
qa/done
mftoure
changed the title
pkg/security/probe/probe_ebpf.go
Outdated
| var errResolution *path.ErrPathResolution | ||
| if errors.As(err, &errResolution) { | ||
| event.SetPathResolutionError(&event.ProcessCacheEntry.FileEvent, err) | ||
| } else { | ||
| seclog.Errorf("failed to insert exec event: %s (offset %d, len %d)", err, offset, len(data)) |
There was a problem hiding this comment.
I think we should report the PID in the error message, it could help to debug
pkg/security/probe/probe_ebpf.go
Outdated
|
|
||
| var exists bool | ||
| event.ProcessCacheEntry, exists = p.fieldHandlers.GetProcessCacheEntry(event, newEntryCb) | ||
| exists := p.Resolvers.ProcessResolver.AddExitEntry(event, newEntryCb) |
There was a problem hiding this comment.
I would rather call it ApplyExitEntry or something similar because in fact we don't add an entry to this process tree, we just update this entry
| p.ApplyBootTime(event.ProcessCacheEntry) | ||
| event.ProcessCacheEntry.SetSpan(event.SpanContext.SpanID, event.SpanContext.TraceID) | ||
|
|
||
| if event.ProcessCacheEntry.Pid == 0 { |
There was a problem hiding this comment.
We could move this at the very beginning of the function ?
There was a problem hiding this comment.
Maybe we should return an error as we do for the AddExecEntry
pkg/security/probe/probe_ebpf.go
Outdated
| seclog.Debugf("failed to resolve new process cache entry context for pid %d: %s", event.PIDContext.Pid, err) | ||
|
|
||
| err = p.Resolvers.ProcessResolver.AddExecEntry(event) | ||
| if err != nil { | ||
| var errResolution *path.ErrPathResolution | ||
| if errors.As(err, &errResolution) { |
There was a problem hiding this comment.
We should move this error handling withing the AddExecEntry function otherwise we are going to change the current behavior which is
Path error => set the event as in error & insert the entry
With your changes, if there is an path resolution error, the entry isn't inserted
| if entry.Pid != 1 { | ||
| parent := p.entryCache[entry.PPid] | ||
| if parent != nil { | ||
| fmt.Println(parent.FileEvent.Inode, inode) |
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=51550885 --os-family=ubuntuNote: This applies to commit 38d853d |
|
/merge |
Devflow running:
|
What does this PR do?
This PR refactors the process resolver tests.
Motivation
Describe how you validated your changes
Possible Drawbacks / Trade-offs
Additional Notes