Files
windmill/benchmarks/sim/toxiproxy.ts
pyranota 74b662d8de feat(benchmarks): k8s sim mode + util-group dashboard + reliability fixes
Stand up a minikube-backed simulation subsystem for benching Windmill under
realistic multi-node load, with a per-bench measurement pipeline and a
dashboard renderer that consolidates throughput, queue depth, per-node CPU,
PG latency/conns, OOM events, and per-node CPU-util-vs-oversaturation into
one SVG report.

Sim infrastructure (sim/):
- k8s_provisioner: minikube up + heterogeneous node sizing from topology JSON
- helm_deploy: helm install Windmill with smoke.yaml + local.yaml overlays
- image_cache: pre-load required images so bench bringup is offline-safe
- toxiproxy_k8s: per-node toxiproxy DaemonSet for cross-node latency injection
- cpu_sampler_k8s: privileged DS reading per-cgroup cpu.stat at 10Hz, dual-
  writes to stdout AND a host-mounted log file (/var/log/wm-sim-cpu-sampler/
  sampler.tsv) so heavy benches no longer lose early samples to kubelet log
  rotation
- pg_logging: ALTER SYSTEM + SIGHUP to enable verbose PG logging without restart
- pgbadger: post-bench PG log analysis HTML report
- readiness: pre-bench cluster health check (samplers stable ≥30s, workers
  ready, PG responsive, queue empty, **deploy.status rollout-complete**) —
  the rollout-complete check catches mid-rolling-update fires that previously
  starved m04's sampler under cgroup_mutex contention

Per-bench JSONL pollers, started/finalized alongside the bench loop:
- pod_timeline: 1Hz workers-per-node Ready counts (used for the workers panel)
- oom_poller: live OOM event capture (kernel + kubelet evictions + cgroup)
- pg_latency_poller: 4Hz psql \\timing on SELECT 1 vs kubectl-exec roundtrip
- pg_conn_poller: 1Hz pg_stat_activity by state (active/idle/idle_in_xact)
- node_load_poller: 2Hz /proc/loadavg + /proc/stat procs_running per node

Dashboard renderer (sim/render_report.ts + graph.ts):
- Util group: one panel per node with translucent orange oversaturation area
  BEHIND solid blue CPU-util area, 100% reference line, phase-boundary verticals.
  cols:2 grid wraps after 2 panels per row.
- PG node tinted with [PG] flag in legend across the dashboard.
- Phase-boundary verticals + push-window shaded zones layered consistently.
- All x-axes switched from wall-clock HH:MM to relative seconds-from-bench-
  start. Shared origin sourced from meta.json's bench_start_ms so 0s on every
  panel = the same wall-clock moment (previously each chart picked its own
  earliest sample as origin, causing drift between panels).

Oversaturation metric, with explicit fallback:
- Primary: (procs_running - ncpu) / ncpu × 100 — true CPU run-queue pressure.
- Fallback to load1 when procs_running is missing (older reports).
- load1 overcounted previously because it includes uninterruptible D-state
  procs (PG backends in disk I/O, cgroup_mutex waits), inflating "saturation"
  by 5-10x under load.
- Pure helper extracted to sim/util_metrics.ts; 8 unit tests cover the
  procs_running > load1 preference, the clamp-at-zero, invalid-ncpu cases.

Sampler reliability:
- HostPath log file in addition to stdout so the bench's scp-based collector
  bypasses kubelet log rotation entirely.
- main.ts truncates the host log file on every node before pushers start
  (parallel ssh, best-effort) so it doesn't grow unbounded across runs.
- Collector falls back to kubectl-logs when scp fails for any node.

Workloads (workloads/):
- io_4phase: four-phase IO step (idle → 2.5s → 500ms → 150ms jobs)
- io_150ms_flood / io_300ms_flood / io_1s_flood / io_2s_flood: single-phase
  flood configs to isolate the worker-host CFS context-switch storm vs PG
  contention regime
- burst, ops_day, cpu_*, etc. for other scenarios

Tests:
- sim/util_metrics_test.ts — 8 cases for computeOversatPct
- sim/util_panel_snapshot_test.ts — 5 assertions guarding util-panel SVG
  invariants (orange behind blue, 100% ref line, relative-time ticks NOT
  wall-clock, phase-boundary verticals, shared-origin override)

Helm values:
- sim/values/smoke.yaml — bench-tuned: workers w/ no CPU limit & low mem
  request, PG w/ 3-core request + wm-critical priorityClass + oomImmune +
  maxConnections, app w/ wm-critical + oomImmune + no resource limits.
- sim/values/local.example.yaml — template for the gitignored local.yaml
  that carries the EE license key.
- Depends on the wm-critical PriorityClass + oomImmune + maxConnections
  knobs landing in windmill-helm-charts (separate PR).

graph.ts additions:
- areaFills param: ordered list of per-kind translucent area fills drawn
  before lines, used by the util panel for orange-behind-blue layering
- lineColorOverrides: pin per-kind line colors so oversaturation reliably
  renders orange regardless of d3 ordinal-color insertion order
- highlightKindToken: substring-match flag for the PG-node tint in Node CPU
- xRelativeOriginMs: shared bench-start origin for the relative-time x-axis
- DataPointMulti is now exported for downstream tests

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-06-08 11:43:47 +02:00

114 lines
3.8 KiB
TypeScript

// Toxiproxy admin client + per-node proxy planning.
//
// One toxiproxy container is run for the whole sim. For every node that
// declares db_latency_ms > 0, we create a proxy with a unique listener inside
// the toxiproxy container; the node's DATABASE_URL points at
// `toxiproxy:<port>` instead of `postgres:5432`. The provisioner is
// responsible for starting the container; this module only configures it.
//
// Wire format docs: https://github.com/Shopify/toxiproxy#http-api
export type ProxyPlan = {
proxyName: string; // toxiproxy proxy id, e.g. "node-far"
listenPort: number; // port inside the toxiproxy container
upstream: string; // e.g. "pg:5432"
latencyMs: number; // 0 means no toxic, plain pass-through
};
// Allocate ports starting at this offset so they don't collide with toxiproxy's
// admin port (8474) or anything else common. 15400 is unused in well-known
// service registries.
export const FIRST_PROXY_PORT = 15400;
export function buildProxyPlans(
nodes: { id: string; db_latency_ms?: number }[],
upstream: string,
): ProxyPlan[] {
const plans: ProxyPlan[] = [];
for (let i = 0; i < nodes.length; i++) {
const n = nodes[i];
if (!n.db_latency_ms || n.db_latency_ms <= 0) continue;
plans.push({
proxyName: `node-${n.id}`,
listenPort: FIRST_PROXY_PORT + i,
upstream,
latencyMs: n.db_latency_ms,
});
}
return plans;
}
export class ToxiproxyClient {
constructor(public adminUrl: string) {}
async waitReady(timeoutMs = 30_000): Promise<void> {
const start = Date.now();
while (Date.now() - start < timeoutMs) {
try {
const r = await fetch(`${this.adminUrl}/version`);
if (r.ok) {
await r.body?.cancel();
return;
}
await r.body?.cancel();
} catch (_) { /* not ready yet */ }
await new Promise((r) => setTimeout(r, 250));
}
throw new Error(`toxiproxy admin API not reachable at ${this.adminUrl}`);
}
async reset(): Promise<void> {
const r = await fetch(`${this.adminUrl}/reset`, { method: "POST" });
if (!r.ok) throw new Error(`toxiproxy reset failed: ${r.status} ${await r.text()}`);
await r.body?.cancel();
}
async createProxy(plan: ProxyPlan): Promise<void> {
const r = await fetch(`${this.adminUrl}/proxies`, {
method: "POST",
headers: { "content-type": "application/json" },
body: JSON.stringify({
name: plan.proxyName,
listen: `0.0.0.0:${plan.listenPort}`,
upstream: plan.upstream,
enabled: true,
}),
});
if (!r.ok) {
const body = await r.text();
throw new Error(`createProxy(${plan.proxyName}) failed: ${r.status} ${body}`);
}
await r.body?.cancel();
}
async addLatencyToxic(proxyName: string, latencyMs: number): Promise<void> {
const r = await fetch(`${this.adminUrl}/proxies/${proxyName}/toxics`, {
method: "POST",
headers: { "content-type": "application/json" },
body: JSON.stringify({
// The default downstream direction adds latency to traffic going from
// upstream (PG) to client (worker), which is what matters for query RTT.
type: "latency",
attributes: { latency: latencyMs, jitter: 0 },
}),
});
if (!r.ok) {
const body = await r.text();
throw new Error(`addLatencyToxic(${proxyName}) failed: ${r.status} ${body}`);
}
await r.body?.cancel();
}
// Convenience: clear all proxies and recreate the plan. Idempotent across
// sim runs that reuse a long-lived toxiproxy container.
async apply(plans: ProxyPlan[]): Promise<void> {
await this.reset();
for (const p of plans) {
await this.createProxy(p);
if (p.latencyMs > 0) {
await this.addLatencyToxic(p.proxyName, p.latencyMs);
}
}
}
}