import hashlib import json import math import re import subprocess import sys import xml.etree.ElementTree as ET from datetime import datetime, timezone, timedelta from pathlib import Path ROOT = Path(__file__).resolve().parents[1] sys.path.insert(0, str(ROOT / "tests")) from pipeline_support import compose, sql, broker, ready_api def main(): ready_api() stamp = datetime.now(timezone(timedelta(hours=3))).isoformat() ids = compose("ps", "-q").splitlines() inspection = json.loads(subprocess.check_output(["docker", "inspect", *ids], text=True)) allowed = { "ROLE", "JOB_MAX_ATTEMPTS", "JOB_TIMEOUT_SECONDS", "JOB_LEASE_SECONDS", "MAX_PENDING_JOBS", "FAULT_INJECTION_ENABLED", } containers = [] for item in inspection: environment = { v.split("=", 1)[0]: v.split("=", 1)[1] for v in item["Config"]["Env"] if v.split("=", 1)[0] in allowed } containers.append( { "name": item["Name"], "image": item["Image"], "started_at": item["State"]["StartedAt"], "state": item["State"]["Status"], "health": item["State"].get("Health", {}).get("Status"), "volumes": [ {"name": m["Name"], "target": m["Destination"]} for m in item["Mounts"] if m["Type"] == "volume" ], "memory_limit_bytes": item["HostConfig"]["Memory"], "nano_cpus": item["HostConfig"]["NanoCpus"], "profile": environment, } ) info = { "time_moscow": stamp, "base_commit": subprocess.check_output(["git", "rev-parse", "HEAD"], text=True).strip(), "source_lab4": "cc0ef71", "worktree": "modified; source hashes below", "docker": subprocess.check_output( ["docker", "info", "--format", "{{.ServerVersion}} CPU={{.NCPU}} RAM={{.MemTotal}}"], text=True, ).strip(), "containers": containers, "job_states": sql( "select status,count(*) from processing_jobs group by status order by status;" ), "redis_durability": broker("CONFIG", "GET", "appendonly", "appendfsync"), "broker_stream_entries": broker("XLEN", "attachment-jobs"), "broker_publication_index_entries": broker("HLEN", "job-publications"), } (ROOT / "evidence/environment.json").write_text(json.dumps(info, ensure_ascii=False, indent=2)) capture_schema() capture_metrics() capture_final_checks() files = list((ROOT / "src").rglob("*.java")) + list( (ROOT / "src/main/resources").rglob("*.sql") ) files += ( list((ROOT / "worker").rglob("*.py")) + list((ROOT / "tests").glob("*.py")) + list((ROOT / "scripts").glob("*")) ) files += [ ROOT / p for p in [ "build.gradle", "settings.gradle", ".editorconfig", "pyproject.toml", "requirements-tools.txt", "gradle/wrapper/gradle-wrapper.properties", "Dockerfile", "worker/Dockerfile", "worker/requirements.txt", "docker-compose.yml", "compose.test.yaml", "compose.local.yaml", "nginx.conf", "policy.json", ] ] manifest = { p.relative_to(ROOT).as_posix(): hashlib.sha256(p.read_bytes()).hexdigest() for p in files if p.is_file() } (ROOT / "evidence/source-sha256.json").write_text(json.dumps(manifest, indent=2)) summary = [ {k: v for k, v in ET.parse(p).getroot().attrib.items() if k != "hostname"} for p in (ROOT / "build/test-results/test").glob("TEST-*.xml") ] (ROOT / "evidence/unit-summary.json").write_text(json.dumps(summary, indent=2)) targets = re.findall(r"\]\(([^)]+)\)", (ROOT / "REPORT.md").read_text()) missing = [t for t in targets if not t.startswith(("http", "#")) and not (ROOT / t).exists()] if missing: raise AssertionError("Missing report links: " + str(missing)) print("PASS environment, job schema, metrics, source hashes and report links") def capture_schema(): from jsonschema import Draft202012Validator, FormatChecker schema = json.loads((ROOT / "contracts/jobs.schema.json").read_text()) validator = Draft202012Validator(schema, format_checker=FormatChecker()) payloads = json.loads( sql("select coalesce(json_agg(payload),'[]'::json) from processing_outbox;") ) for payload in payloads: validator.validate(payload) (ROOT / "evidence/job-schema.json").write_text( json.dumps({"validated_outbox_templates": len(payloads), "valid": True}, indent=2) ) def capture_final_checks(): public = [ "README.md", "lab.json", "contracts", "tests/client.py", "tests/smoke.py", "tests/resilience.py", "tests/load.py", "tests/image_probe.py", ] subprocess.run(["git", "diff", "--exit-code", "--", *public], cwd=ROOT, check=True) subprocess.run(["git", "diff", "--check"], cwd=ROOT, check=True) names = ( subprocess.check_output( ["git", "ls-files", "--cached", "--others", "--exclude-standard", "-z"], cwd=ROOT ) .decode("utf-8") .split("\0") ) credentials = [ path.read_bytes().strip() for path in (ROOT / "secrets").glob("*.txt") if "password" in path.stem or "secret" in path.stem ] for name in filter(None, names): path = ROOT / name if path.is_file() and any(value and value in path.read_bytes() for value in credentials): raise AssertionError("Local credential value found in a Git-visible file") subprocess.run(["git", "check-ignore", "--quiet", ".env"], cwd=ROOT, check=True) for path in (ROOT / "secrets").glob("*.txt"): subprocess.run(["git", "check-ignore", "--quiet", str(path)], cwd=ROOT, check=True) (ROOT / "evidence/final-checks.txt").write_text( "PASS public lab5 assignment, contract and executable tests unchanged\n" "PASS no local credential values in Git-visible files; secrets and .env ignored\n" "PASS git diff --check\n" "Controller DI, single service call, API Javadoc and formatting: evidence/style.txt\n" ) def capture_metrics(): raw = json.loads((ROOT / "evidence/backpressure.json").read_text()) peaks = {} for sample in raw["stats_samples"]: for name, value, unit, cpu in re.findall( r"(\S+) ([\d.]+)(KiB|MiB|GiB) / \S+ ([\d.]+)%", sample ): value = float(value) * {"KiB": 1 / 1024, "MiB": 1, "GiB": 1024}[unit] current = peaks.setdefault(name, {"sampled_peak_MiB": 0, "sampled_peak_CPU_percent": 0}) current["sampled_peak_MiB"] = max(current["sampled_peak_MiB"], value) current["sampled_peak_CPU_percent"] = max( current["sampled_peak_CPU_percent"], float(cpu) ) result = { "profile": "6 accepted / 4 unavailable; 10 clients; 800x600 PNG", "limits": "API 512MiB, worker 1GiB, dispatcher 256MiB", "samples": len(raw["stats_samples"]), "peaks": peaks, "limitations": "discrete RSS/CPU samples, not true maximum or steady-state throughput", } (ROOT / "evidence/resource-summary.json").write_text(json.dumps(result, indent=2)) observations = sorted(raw["seconds_to_observed_ready"]) drain = { "observed_ready_seconds": { str(p): observations[max(0, math.ceil(p / 100 * len(observations)) - 1)] for p in (50, 95, 99) }, "accepted": raw["accepted"], "expected_unavailable": raw["rejected"], "image_bytes": raw["image_bytes"], "polling_and_startup_included": True, "interpretation": "upper bounds from sequential polls, n=6; not a steady-state latency SLA", } (ROOT / "evidence/drain-summary.json").write_text(json.dumps(drain, indent=2)) if __name__ == "__main__": main()