Files
windmill/benchmarks/sim/node_load_poller.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

106 lines
4.2 KiB
TypeScript

// Polls /proc/loadavg + nproc on each minikube node every intervalMs and
// writes one JSONL row per node per tick. Used to compute *saturation* (load /
// ncpu) and *oversaturation* (max(0, load/ncpu - 1)) in the dashboard. PSI is
// not available in the minikube kernel and cpu.stat throttling is meaningless
// without limits.cpu, so loadavg is the only saturation signal we have.
//
// Output line: {"ts": ms, "node": "wm-sim-k8s-4node-m02", "load1": 44.2,
// "load5": 32.1, "load15": 24.4, "ncpu": 4}
import { MinikubeProvisioner } from "./k8s_provisioner.ts";
export type NodeLoadPoller = {
cont: { value: boolean };
done: Promise<void>;
};
export function startNodeLoadPoller(
prov: MinikubeProvisioner,
outPath: string,
opts: { intervalMs?: number } = {},
): NodeLoadPoller {
const intervalMs = opts.intervalMs ?? 2000;
const cont = { value: true };
const f = Deno.openSync(outPath, { write: true, create: true, truncate: true });
const enc = new TextEncoder();
const done = (async () => {
// Discover nodes once at startup. The poll loop reuses this list. New
// nodes joining mid-bench are rare (we don't auto-scale the cluster).
let nodes: { name: string; ip: string }[] = [];
try {
const r = await prov.kubectl([
"get", "nodes",
"-o", "jsonpath={range .items[*]}{.metadata.name}{\"|\"}{.status.addresses[?(@.type==\"InternalIP\")].address}{\"\\n\"}{end}",
]);
if (r.code === 0) {
for (const line of r.stdout.split("\n")) {
if (!line.trim()) continue;
const [name, ip] = line.split("|");
if (name && ip) nodes.push({ name: name.trim(), ip: ip.trim() });
}
}
} catch (e) {
console.warn(`[node-load] node discovery failed: ${(e as Error).message}`);
}
while (cont.value) {
const startMs = Date.now();
// Poll all nodes in parallel — one ssh per node per tick.
await Promise.all(nodes.map(async ({ name, ip }) => {
try {
// Each minikube node has its own ssh key under ~/.minikube/machines.
const keyPath = `${Deno.env.get("HOME")}/.minikube/machines/${name}/id_rsa`;
const proc = new Deno.Command("ssh", {
args: [
"-o", "StrictHostKeyChecking=no",
"-o", "UserKnownHostsFile=/dev/null",
"-o", "ConnectTimeout=2",
"-o", "LogLevel=ERROR",
"-i", keyPath,
`docker@${ip}`,
// procs_running is the runnable count (CPU-bound queue) — does
// NOT include D-state procs (disk/network wait). loadavg counts
// both, so loadavg/ncpu was conflating CPU-starved processes
// with PG backends waiting on disk I/O.
"cat /proc/loadavg && cat /proc/stat | grep ^procs_running && nproc",
],
stdout: "piped",
stderr: "null",
});
const out = await proc.output();
const text = new TextDecoder().decode(out.stdout).trim();
const lines = text.split("\n");
if (lines.length < 3) return;
const loadParts = lines[0].split(" ");
const load1 = parseFloat(loadParts[0]);
const load5 = parseFloat(loadParts[1]);
const load15 = parseFloat(loadParts[2]);
// "procs_running N" — instantaneous count of runnable processes
// (current + queued for CPU). Excludes D-state.
const procsRunning = parseInt(lines[1].split(/\s+/)[1] ?? "");
const ncpu = parseInt(lines[2].trim());
if (!Number.isFinite(load1) || !Number.isFinite(ncpu)) return;
const row = {
ts: startMs,
node: name,
load1,
load5,
load15,
procs_running: Number.isFinite(procsRunning) ? procsRunning : null,
ncpu,
};
f.writeSync(enc.encode(JSON.stringify(row) + "\n"));
} catch (_e) { /* skip this node this tick */ }
}));
const elapsed = Date.now() - startMs;
if (cont.value && elapsed < intervalMs) {
await new Promise((r) => setTimeout(r, intervalMs - elapsed));
}
}
f.close();
})();
return { cont, done };
}