vumast3r/cka-sim
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
43e2e1db2f
cka-sim/exam
vumast3r 43e2e1db2f refactor: renumber the CKA bank 1-102 with no gaps 
Task ids used to jump 1-43 / 101-124 / 401-418 / 501-517. They now run
contiguously in the same order:

  1-43    unchanged (base and extra)
  44-67   hard pool          (was 101-124)
  68-85   Simulator Set C    (was 401-418)
  86-102  Simulator Set D    (was 501-517)

Every reference moved with its task: ids, solution keys, workspace paths
({WS}/qNN and the graders' hidden .qNN-* files), temp paths, the one
hardcoded id in code (Q56's derived canary weight), and Q-number
cross-references in the harness, README and the explainer. Each task's
paths were checked to carry only its own new number. History entries and
the local session state were remapped so past attempts stay attached to the
same tasks; retired set A/B ids keep their old numbers and match nothing.
weak-spots is regenerated.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mhxm4rsUfuE5HoytYbu2YW
2026-09-22 22:05:37 -06:00

1388 lines
71 KiB
Python
Executable File
Raw Blame History

#!/usr/bin/env python3
"""exam - Kubernetes certification exam simulator (CKA / CKAD / CKS tracks).
Hands-on tasks graded per-check against the live cluster, exam clock, attempt history,
weak-spot log, drill mode, and a browser front end modelled on the real exam."""
import http.server
import importlib
import json
import os
import shutil
import signal
import subprocess
import sys
import threading
import time
import urllib.parse
from collections.abc import Callable
from datetime import datetime
from typing import Any
VERSION = "2026.09.07.1"
ROOT = os.path.dirname(os.path.abspath(__file__))
sys.path.insert(0, ROOT)
# ------------------------------------------------------------ track ----
ARGV = sys.argv[1:]
TRACKS = ["cka", "ckad", "cks"]
TRACK_ID = ""
TRACK = {}
QUESTIONS = []
DOMAINS = {}
QUOTA = {}
QUOTA_HARD = {}
QBYID = {}
SETS = {}
POOL = []
EXAM_SECONDS = 0
PASS_PCT = 0
STATE_FILE = ""
HISTORY_FILE = ""
WEAK_FILE = ""
def load_track(track_id):
"""Bind every track-dependent global. Called at startup and when the UI switches track."""
try:
mod = importlib.import_module(f"tracks.{track_id}.questions")
except ModuleNotFoundError:
print(f"unknown track '{track_id}' (no tracks/{track_id}/questions.py)"); sys.exit(1)
importlib.reload(mod)
track, questions = mod.TRACK, list(mod.QUESTIONS)
for q in questions:
q["weight"] = sum(c["pts"] for c in q["checks"])
q.setdefault("difficulty", "normal")
q.setdefault("set", None)
g = globals()
g.update(
TRACK_ID=track_id, TRACK=track, QUESTIONS=questions, DOMAINS=track["domains"],
QUOTA=track.get("quota", {}), QUOTA_HARD=track.get("quota_hard", track.get("quota", {})),
QBYID={q["id"]: q for q in questions},
SETS=track.get("sets", {}),
POOL=[q for q in questions if not q.get("set")],
EXAM_SECONDS=track["minutes"] * 60, PASS_PCT=track["pass_pct"],
STATE_FILE=os.path.join(ROOT, ".state", f"state-{track_id}.json"),
HISTORY_FILE=os.path.join(ROOT, f"history-{track_id}.jsonl"),
WEAK_FILE=os.path.join(ROOT, f"weak-spots-{track_id}.md"),
)
def set_questions(name):
return [q for q in QUESTIONS if q.get("set") == name]
STATE_DIR = os.path.join(ROOT, ".state")
WORKSPACE = os.path.join(ROOT, "workspace")
# How {WS} is shown in task text and solutions. Relative to the repo root by default, since that
# is where you are working; graders always use the absolute $WS internally.
WS_DISPLAY = os.environ.get("CKA_WS_DISPLAY") or os.path.relpath(WORKSPACE, ROOT)
DRILL_SECONDS_PER_TASK = 6 * 60
DRILL_SECONDS_PER_TASK_HARD = 5 * 60
DRILL_SIZE = 5
import random
NAMESPACES_TO_SETTLE = ["apps", "netpol", "svc", "ingress-app", "dns-check", "storage"]
# --- cluster targeting (override with env vars to use a non-kind cluster) ---
CONTEXT = os.environ.get("CKA_CONTEXT", "kind-cka-sim")
def _default_access():
try:
return open(os.path.join(STATE_DIR, "runtime")).read().strip() # written by setup.sh
except OSError:
return "podman"
NODE_ACCESS = os.environ.get("CKA_NODE_ACCESS", _default_access()) # podman | docker | ssh | vagrant
NODES = {
"CP": os.environ.get("CKA_CP", "cka-sim-control-plane"),
"W1": os.environ.get("CKA_W1", "cka-sim-worker"),
"W2": os.environ.get("CKA_W2", "cka-sim-worker2"),
"W3": os.environ.get("CKA_W3", "cka-sim-worker3"),
}
def shell_hint(node):
return {"podman": f"podman exec -it {node} bash",
"docker": f"docker exec -it {node} bash",
"ssh": f"ssh {node}",
"vagrant": f"vagrant ssh {node}"}.get(NODE_ACCESS, f"ssh {node}")
PRELUDE = r"""
set -uo pipefail
K="kubectl --context ${CKA_CONTEXT:-kind-cka-sim}" # grading never depends on the candidate's current context
WS="$WORKSPACE"
# j [kubectl get args...] '{jsonpath}' -> prints the jsonpath result ("" on error)
j() { local n=$#; local path="${!n}"; $K get "${@:1:n-1}" -o "jsonpath=$path" 2>/dev/null; }
eq() { [ "$1" = "$2" ]; }
CP="${CKA_CP:-cka-sim-control-plane}"; W1="${CKA_W1:-cka-sim-worker}"
W2="${CKA_W2:-cka-sim-worker2}"; W3="${CKA_W3:-cka-sim-worker3}"
# nx NODE 'command' -> run a shell command on a node (stdin is forwarded)
nx() {
local n="$1"; shift
case "${CKA_NODE_ACCESS:-podman}" in
ssh) ssh -o BatchMode=yes -o StrictHostKeyChecking=accept-new "$n" "$@" ;;
vagrant) vagrant ssh "$n" -c "$*" -- -o LogLevel=QUIET ;;
docker) docker exec -i "$n" sh -c "$*" ;;
*) podman exec -i "$n" sh -c "$*" ;;
esac
}
# ensure_etcd_tools -> etcdctl and etcdutl on the control-plane node, as on the real exam.
# The etcd image is distroless, so they cannot be copied out of the pod; etcd-tools.sh takes
# them from the image in the node's containerd (and falls back to a download / apt).
ensure_etcd_tools() {
nx "$CP" 'command -v etcdctl >/dev/null 2>&1 && command -v etcdutl >/dev/null 2>&1' </dev/null && return 0
bash "${CKA_ROOT:-.}/etcd-tools.sh" --emit | nx "$CP" 'sh -s'
}
"""
def _pick_track_arg(argv):
t = os.environ.get("CKA_TRACK", "cka")
if "--track" in argv:
i = argv.index("--track"); t = argv[i + 1]; del argv[i:i + 2]
return t
load_track(_pick_track_arg(ARGV))
TTY = sys.stdout.isatty()
def c(code, s): return f"\033[{code}m{s}\033[0m" if TTY else s
def green(s): return c("32", s)
def red(s): return c("31", s)
def yellow(s): return c("33", s)
def bold(s): return c("1", s)
def dim(s): return c("2", s)
def die(msg, code=1):
print(red(msg)); sys.exit(code)
# ------------------------------------------------------------- state ----
def load_state() -> dict[str, Any] | None:
if not os.path.exists(STATE_FILE):
return None
with open(STATE_FILE) as f:
return json.load(f)
def save_state(st: dict[str, Any]) -> None:
os.makedirs(STATE_DIR, exist_ok=True)
with open(STATE_FILE, "w") as f:
json.dump(st, f, indent=2)
def paused_total(st: dict[str, Any] | None) -> float:
"""Seconds spent paused so far, including the current pause if any."""
if not st:
return 0.0
t = float(st.get("paused_total", 0))
if st.get("paused_at"):
t += time.time() - float(st["paused_at"])
return t
def untimed(st: dict[str, Any] | None) -> bool:
return bool(st and st.get("untimed"))
def remaining(st: dict[str, Any] | None) -> float:
if not st:
return 0.0
if untimed(st):
return float("inf")
return float(st.get("started_at", 0)) + float(st.get("duration", EXAM_SECONDS)) + paused_total(st) - time.time()
def set_paused(st: dict[str, Any], on: bool) -> None:
if on and not st.get("paused_at"):
st["paused_at"] = time.time()
elif not on and st.get("paused_at"):
st["paused_total"] = float(st.get("paused_total", 0)) + (time.time() - float(st["paused_at"])); st["paused_at"] = None
save_state(st)
def fmt(secs: float | int) -> str:
secs_int = max(0, int(secs))
return f"{secs_int // 3600:02d}:{(secs_int % 3600) // 60:02d}:{secs_int % 60:02d}"
def require_started() -> dict[str, Any]:
st = load_state()
if not st:
die("No session in progress. Run: ./exam start (or ./exam drill weak)")
assert st is not None
return st
def session_questions(st: dict[str, Any] | None) -> list[dict[str, Any]]:
if not st:
return list(QUESTIONS)
ids = st.get("ids") or [q["id"] for q in QUESTIONS]
# Preserve the session's own order. Filtering QUESTIONS (which is sorted by id) threw it away,
# so compose_exam's rng.shuffle never reached the candidate and every exam arrived grouped by
# domain and difficulty - base tasks first, then hard, then the sim sets. Real exams are not
# ordered by difficulty, and practising triage on a sorted list teaches the wrong instinct.
by_id = {q["id"]: q for q in QUESTIONS}
return [by_id[i] for i in ids if i in by_id]
def q_by_id(n: int) -> dict[str, Any]:
for q in QUESTIONS:
if q["id"] == n:
return q
# ids run 1..N without gaps today, but print the real ranges rather than "1-<count>" so a
# future gap can never make this message name ids that do not exist.
ids = sorted(q["id"] for q in QUESTIONS)
ranges, start, prev = [], ids[0], ids[0]
for i in ids[1:] + [None]:
if i != prev + 1:
ranges.append(str(start) if start == prev else f"{start}-{prev}")
start = i
prev = i
die(f"No question {n}. Valid: {', '.join(ranges)}")
return {}
def render(text, params):
"""Substitute %{param}, {WS}, {CP}.. and {SHELL_CP}.. placeholders."""
t = text.replace("{WS}", WS_DISPLAY)
for k, v in (params or {}).items():
t = t.replace("%{" + k + "}", str(v))
for k, v in NODES.items():
t = t.replace("{" + k + "}", v).replace("{SHELL_" + k + "}", shell_hint(v))
return t
def session_params(st, q):
"""Params drawn for this session; outside a session, a fixed sample so text still reads naturally."""
p = (st or {}).get("params", {}).get(str(q["id"]))
return p if p is not None else make_params(q, random.Random(q["id"]))
def task_text(q, params=None):
return render(q["task"].strip(), params if params is not None else session_params(load_state(), q))
def make_params(q, rng):
fn = q.get("params")
return fn(rng) if fn else {}
# -------------------------------------------------------------- shell ----
def run(script, timeout=180, params=None):
env = dict(os.environ, WORKSPACE=WORKSPACE, CKA_ROOT=ROOT, CKA_NODE_ACCESS=NODE_ACCESS, CKA_CONTEXT=CONTEXT,
**{f"CKA_{k}": v for k, v in NODES.items()})
p = subprocess.run(["bash", "-c", PRELUDE + render(script, params)], env=env, capture_output=True, text=True,
timeout=timeout, stdin=subprocess.DEVNULL)
return p.returncode, (p.stdout + p.stderr).strip()
def kubectl_ok():
try:
return subprocess.run(["kubectl", "--context", CONTEXT, "get", "--raw=/version"],
capture_output=True, text=True, timeout=15).returncode == 0
except Exception:
return False
def grade_one(q, params=None):
"""Run every check independently. Returns a result dict with partial credit."""
if params is None:
params = session_params(load_state(), q)
checks = []
for ch in q["checks"]:
try:
rc, out = run(ch["script"], timeout=240, params=params)
except subprocess.TimeoutExpired:
rc, out = 1, "timed out"
checks.append({"name": render(ch["name"], params), "pts": ch["pts"], "pass": rc == 0, "out": out[-400:]})
earned = sum(ch["pts"] for ch in checks if ch["pass"])
return {"pass": earned == q["weight"], "earned": earned, "total": q["weight"],
"checks": checks, "at": time.time(), "difficulty": q["difficulty"]}
def result_detail(r):
return "\n".join((" ✓ " if ch["pass"] else " ✗ ") + ch["name"] + f" ({ch['pts']} pt)" for ch in r["checks"])
def score(st, results=None):
results = results if results is not None else st["results"]
qs = session_questions(st)
total = sum(q["weight"] for q in qs)
earned = sum(results.get(str(q["id"]), {}).get("earned", 0) for q in qs)
return earned, total, (round(100 * earned / total, 1) if total else 0)
# ----------------------------------------------------------- history ----
def load_history(include_synthetic=False):
"""Attempts from history-<track>.jsonl. Entries marked "synthetic" (a scripted verification
run, not a person sitting the tasks) stay in the file for the record but are left out of every
statistic, weak list and "latest exam" unless explicitly asked for."""
if not os.path.exists(HISTORY_FILE):
return []
with open(HISTORY_FILE) as f:
entries = [json.loads(l) for l in f if l.strip()]
return entries if include_synthetic else [e for e in entries if not e.get("synthetic")]
def record_attempt(st):
earned, total, pct = score(st)
entry = {"at": datetime.now().isoformat(timespec="minutes"), "mode": st.get("mode", "exam"), "hard": st.get("hard", False),
"ids": [q["id"] for q in session_questions(st)], "score_pct": pct, "earned": earned, "total": total,
"time_used": int(st["ended_at"] - st["started_at"] - paused_total(st)),
"results": {k: {"pass": v["pass"], "earned": v["earned"], "total": v["total"],
"failed": [ch["name"] for ch in v["checks"] if not ch["pass"]]}
for k, v in st["results"].items()}}
with open(HISTORY_FILE, "a") as f:
f.write(json.dumps(entry) + "\n")
write_weak_spots()
return entry
class AggEntry:
def __init__(self) -> None:
self.attempts: int = 0
self.fails: int = 0
self.last: Any = None
self.last_fail_at: str = ""
self.failed: list[str] = []
self.last_earned: str = ""
self.streak: int = 0 # consecutive full passes at the end of the history
def __getitem__(self, item: str) -> Any:
return getattr(self, item)
def __setitem__(self, item: str, value: Any) -> None:
setattr(self, item, value)
def to_dict(self) -> dict[str, Any]:
return {
"attempts": self.attempts,
"fails": self.fails,
"last": self.last,
"last_fail_at": self.last_fail_at,
"failed": self.failed,
"last_earned": self.last_earned,
"streak": self.streak,
}
def stats() -> dict[int, AggEntry]:
agg: dict[int, AggEntry] = {int(q["id"]): AggEntry() for q in QUESTIONS}
for e in load_history():
for k, r in e.get("results", {}).items():
a = agg.get(int(k))
if a is None:
continue
a.attempts += 1
a.last = r.get("pass")
a.last_earned = f"{r.get('earned', 0)}/{r.get('total', 0)}"
a.streak = a.streak + 1 if r.get("pass") else 0
if not r.get("pass"):
a.fails += 1
a.last_fail_at = str(e.get("at", ""))
a.failed = list(r.get("failed", []))
return agg
def last_seen():
"""question id -> index of the most recent session it appeared in (higher = more recent)."""
seen = {}
for i, e in enumerate(load_history()):
for k in e["results"]:
seen[int(k)] = i
return seen
def drilled_domains():
out = {d: 0 for d in DOMAINS}
for e in load_history():
if e["mode"] != "drill":
continue
for k in e["results"]:
q = QBYID.get(int(k))
if q:
out[q["domain"]] += 1
return out
def breaks_of(q):
"""A task's "breaks" as a set. It is a string or a list: Q61 damages the scheduler AND the
controller-manager, and declaring only one let it share a session with Q38."""
b = q.get("breaks")
return set() if not b else ({b} if isinstance(b, str) else set(b))
def no_conflicts(cands, n, taken=()):
"""Take up to n tasks, never two that break the same thing (or clash with tasks already taken).
A "solo" task (Q45's etcd restore rolls the WHOLE cluster back to its snapshot, wiping every
other task's seed and answers) is only ever taken on its own."""
used = set().union(*(breaks_of(q) for q in taken)) if taken else set()
out = []
for q in cands:
if q.get("solo") and (taken or out):
continue
if any(x.get("solo") for x in list(taken) + out):
break
b = breaks_of(q)
if b & used:
continue
out.append(q)
used |= b
if len(out) == n:
break
return out
def split_solo(qs):
"""Pull solo tasks out of a multi-task selection; they have to be drilled alone."""
if len(qs) <= 1:
return list(qs), []
return [q for q in qs if not q.get("solo")], [q for q in qs if q.get("solo")]
def pick(domain, n, hard, rng, exclude=(), taken=(), unseen_only=False):
"""n tasks from a domain: unseen first, then least recently seen; hard sessions prefer hard tasks."""
seen = last_seen()
def build(allow_hard):
pool = [q for q in POOL if q["domain"] == domain and q["id"] not in exclude and not q.get("solo")
and (allow_hard or q["difficulty"] == "normal") and not (unseen_only and q["id"] in seen)]
rng.shuffle(pool)
pool.sort(key=lambda q: (0 if (hard and q["difficulty"] == "hard") else 1, seen.get(q["id"], -1)))
return pool
got = no_conflicts(build(hard), n, taken)
if len(got) < n and not hard:
# A domain whose only tasks are "hard" contributes NOTHING to a normal session and drops
# out of the blueprint entirely - CKS SUPPLY and RUNTIME (20% each) were both unreachable
# via ./exam start. Better to hand over a hard task than to silently skip the domain.
got = no_conflicts(build(True), n, taken)
return got
def compose_exam(hard, rng):
quota = QUOTA_HARD if hard else QUOTA
# Scarcest domain first. Two tasks that "breaks" the same thing can't share a session, so a
# one-task domain loses every such tie to a bigger domain that claimed the shared slot earlier
# and then contributes nothing: CKS RUNTIME (only Q9, apiserver-flag) was never drawn in ANY
# mode because SETUP was iterated first and took that slot with Q2.
order = sorted(quota, key=lambda d: len([q for q in POOL if q["domain"] == d]))
qs, short = [], {}
for d in order:
got = pick(d, quota[d], hard, rng, taken=qs)
if len(got) < quota[d]:
short[d] = (len(got), quota[d])
qs += got
if short:
# Never truncate in silence - a short session still gets scored out of its own total, so
# the missing domains look like they were simply not on the blueprint.
print(yellow(" warning: bank too small to fill the blueprint - "
+ ", ".join(f"{d} {g}/{w}" for d, (g, w) in short.items())))
rng.shuffle(qs)
return qs
def campaign_batch(hard, rng):
"""'drill all': DRILL_SIZE tasks you have never attempted, round-robin across domains,
hard tasks included when --hard. When everything has been seen once, falls back to weak tasks."""
seen = last_seen()
pool = [q for q in POOL if q["id"] not in seen and not q.get("solo") and (hard or q["difficulty"] == "normal")]
if not pool:
return None, []
qs, domains = [], list(DOMAINS)
rng.shuffle(pool)
pool.sort(key=lambda q: 0 if (hard and q["difficulty"] == "hard") else 1)
while len(qs) < DRILL_SIZE and pool:
progressed = False
for d in domains:
cand = [q for q in pool if q["domain"] == d and q not in qs]
got = no_conflicts(cand, 1, qs)
if got:
qs += got; progressed = True
if len(qs) == DRILL_SIZE:
break
if not progressed:
break
total = len([q for q in POOL if hard or q["difficulty"] == "normal"])
return f"{total - len(pool)}/{total} attempted", qs
def solution_allowed(st: dict[str, Any] | None, qid: int | str) -> bool:
"""Study: always. Drill: once the task has been graded. Exam: only after submit."""
if not st or st.get("ended_at") or untimed(st):
return True
return bool(st.get("mode") == "drill" and str(qid) in st.get("results", {}))
def cmd_solution(args):
if not args:
die("usage: ./exam solution N (drills: after grading the task; exams: after submit; --force to peek anyway)")
force = "--force" in args; args = [a for a in args if a != "--force"]
q = q_by_id(int(args[0])); st = load_state()
if not solution_allowed(st, int(q["id"])) and not force:
die("Not yet: in a drill, grade the task first (./exam grade N); in an exam, submit first. --force overrides.")
print(bold(f"Solution - Q{q['id']} {q['title']}")); print("-" * 64)
sol = str(q.get("solution", ""))
print(render(sol, session_params(st, q))); print("-" * 64)
print(dim(f"One correct way, not the only one. $do = --dry-run=client -o yaml, k = kubectl. Paths are relative to {ROOT}"))
def cmd_progress(args):
seen = last_seen(); agg = stats()
print(bold(f"{'Domain':<8}{'attempted':>11}{'passed':>9} {'hard attempted':>15}{'hard passed':>13}"))
for d in DOMAINS:
dq = [q for q in QUESTIONS if q["domain"] == d]
n = [q for q in dq if q["difficulty"] == "normal"]; h = [q for q in dq if q["difficulty"] == "hard"]
att = lambda L: sum(1 for q in L if q["id"] in seen); ok = lambda L: sum(1 for q in L if agg.get(int(q["id"])) and agg[int(q["id"])].last is True)
print(f"{d:<8}{att(n):>6}/{len(n):<4}{ok(n):>6}/{len(n):<4} {att(h):>9}/{len(h):<4}{ok(h):>8}/{len(h):<4}")
left = [q["id"] for q in QUESTIONS if q["id"] not in seen]
print(dim(f"\n{len(QUESTIONS) - len(left)}/{len(QUESTIONS)} tasks attempted at least once." + (f" Never seen: Q{', Q'.join(map(str, left))}" if left and len(left) <= 25 else "")))
print(dim("passed = latest attempt was a full pass. ./exam drill all [--hard] works through what's left."))
def is_weak(q, a):
"""Weak = failed on the latest attempt, or failed at least twice and not yet passed twice in a
row since - a flaky task has to prove itself - or never attempted (outside the fixed sets).
The old rule ("failed twice, ever") kept a task weak forever: 13 of 29 listed were passing."""
if a.attempts == 0:
return not q.get("set")
return a.last is False or (a.fails >= 2 and a.streak < 2)
def weak_ids(agg=None):
agg = agg or stats()
return [q["id"] for q in QUESTIONS if is_weak(q, agg[q["id"]])]
def weak_lines(a):
"""What to show under a weak task: the failing checks if it failed last time, otherwise how
close it is to dropping off the list - not the ✗ list of some older, since-fixed attempt."""
if a.last is False:
return [f"✗ {n}" for n in (a.failed or [])]
return [f"passed last time ({a.last_earned}) - one more clean pass takes it off this list"]
def exam_domain_scores(e):
"""domain -> (earned, total) over the tasks that were actually IN that attempt. Summing the
weight of every task in the bank instead made a 100% run read "ARCH 38/266"."""
out = {}
for k, r in e.get("results", {}).items():
q = QBYID.get(int(k))
if q:
de, dt = out.get(q["domain"], (0, 0))
out[q["domain"]] = (de + r.get("earned", 0), dt + r.get("total", 0))
return out
def write_weak_spots():
agg = stats(); hist = load_history()
L = [f"# {TRACK_ID.upper()} weak spots", "", f"_Generated by exam from {len(hist)} attempt(s); regenerated on every `end`._", ""]
exams = [e for e in hist if e["mode"] == "exam"]
if exams:
e = exams[-1]; ds = exam_domain_scores(e)
L += [f"## Domain scores (latest full exam, {e['at']}, {e['score_pct']}%)", ""]
for d, name in DOMAINS.items():
if d in ds:
L.append(f"- **{d}** {ds[d][0]}/{ds[d][1]} — {name}")
L.append("")
L += ["## Drill next (`./exam drill weak`)", ""]
weak = [q for q in QUESTIONS if q["id"] in weak_ids(agg) and agg[q["id"]]["attempts"] > 0]
if not weak:
L.append("Nothing failing right now.")
for q in weak:
a = agg[q["id"]]
L.append(f"### Q{q['id']:02d} · {q['domain']} · {q['title']} ({a['fails']}/{a['attempts']} failed, last {a['last_earned']} pts, last failed {a['last_fail_at']})")
L += [f"- {n}" for n in weak_lines(a)]
L.append("")
L += ["## Attempt matrix", "", "| Q | Domain | Title | " + " | ".join(f"{e['mode'][0].upper()}{i + 1}" for i, e in enumerate(hist)) + " |",
"|---|---|---|" + "---|" * len(hist)]
for q in QUESTIONS:
cells = []
for e in hist:
r = e["results"].get(str(q["id"]))
cells.append("·" if r is None else ("✓" if r["pass"] else f"{r['earned']}/{r['total']}"))
L.append(f"| {q['id']} | {q['domain']} | {q['title']} | " + " | ".join(cells) + " |")
L += ["", "_E = full exam, D = drill, · = not in that session, n/m = partial credit._", ""]
with open(WEAK_FILE, "w") as f:
f.write("\n".join(L))
# ------------------------------------------------------------ session ----
def preflight():
out = subprocess.run(["kubectl", "--context", CONTEXT, "get", "nodes", "-o", "jsonpath={.items[*].metadata.name}"],
capture_output=True, text=True).stdout.split()
missing = [v for v in NODES.values() if v not in out]
if missing:
die(f"Nodes not found in cluster: {missing}\n cluster has: {out}\n"
" Set CKA_CP, CKA_W1, CKA_W2, CKA_W3 to your node names (needs 1 control-plane + 3 workers).")
rc, _ = run('nx "$CP" "true" </dev/null && nx "$W1" "true" </dev/null && nx "$W2" "true" </dev/null && nx "$W3" "true" </dev/null', timeout=60)
if rc != 0:
die(f"Cannot run commands on nodes with CKA_NODE_ACCESS={NODE_ACCESS}.\n"
" podman/docker -> node names must be kind container names (kind get nodes --name cka-sim)\n"
" ssh -> passwordless ssh to each node name must work\n"
" vagrant -> run from the Vagrantfile directory; names are vagrant machine names")
print(dim(f" track: {TRACK_ID} cluster: {CONTEXT} node access: {NODE_ACCESS} nodes: {', '.join(NODES.values())}"))
for tool, why in (("helm", "Helm tasks (the real exam has helm on the terminal)"), ("openssl", "certificate tasks"), ("curl", "tasks that fetch manifests")):
if not shutil.which(str(tool)):
print(yellow(f" warning: '{tool}' not found on this machine - needed for {why}"))
HARD_CONTEXTS = {"k8s-c1-H": "default", "k8s-c2-AC": "%{ns}", "k8s-c3-CCC": "kube-system"}
def context_namespace(ctx, template, params_by_id):
"""Default namespace for one exam context.
Several tasks can name the same context while working in different namespaces, but a
context has only one default namespace - so it belongs to whichever task actually reads it
("ctx_ns": True, i.e. the task says to work in the context's namespace without -n). Every
other task states its namespace in its own text, so the value does not matter to them.
"""
if "%{ns}" not in template:
return template
def ns_of(pred):
return [p["ns"] for i, p in params_by_id.items()
if p.get("ctx") == ctx and p.get("ns") and pred(QBYID.get(int(i)) or {})]
return (ns_of(lambda q: q.get("ctx_ns")) or ns_of(lambda q: True)
or [p["ns"] for p in params_by_id.values() if p.get("ns")] or ["default"])[0]
def setup_contexts(hard, params_by_id):
"""Exam-style contexts on the same cluster with different default namespaces.
Created whenever a task in the session asks for one - 24 tasks open with
`kubectl config use-context k8s-c?-?` and are unfollowable without it - not only when the
whole session was started with --hard. A drill that names hard task ids explicitly
('./exam drill 121', './exam drill weak') pulls those tasks in with hard=False.
"""
cluster = subprocess.run(["kubectl", "config", "view", "-o", f"jsonpath={{.contexts[?(@.name=='{CONTEXT}')].context.cluster}}"],
capture_output=True, text=True).stdout.strip()
user = subprocess.run(["kubectl", "config", "view", "-o", f"jsonpath={{.contexts[?(@.name=='{CONTEXT}')].context.user}}"],
capture_output=True, text=True).stdout.strip()
for name in HARD_CONTEXTS:
subprocess.run(["kubectl", "config", "delete-context", name], capture_output=True)
wanted = set(HARD_CONTEXTS) if hard else {p.get("ctx") for p in params_by_id.values()} & set(HARD_CONTEXTS)
if not wanted:
subprocess.run(["kubectl", "config", "use-context", CONTEXT], capture_output=True)
return
for name in sorted(wanted):
subprocess.run(["kubectl", "config", "set-context", name, f"--cluster={cluster}", f"--user={user}",
f"--namespace={context_namespace(name, HARD_CONTEXTS[name], params_by_id)}"], capture_output=True)
# Hard mode starts you in c1 like the real exam; a mixed drill stays on the normal context
# so its non-hard tasks behave as usual and the hard ones still switch as instructed.
start = "k8s-c1-H" if hard and "k8s-c1-H" in wanted else CONTEXT
subprocess.run(["kubectl", "config", "use-context", start], capture_output=True)
def _log_msg(*args: Any, **kwargs: Any) -> None:
print(*args, **kwargs)
# Disruptive tasks (see "breaks": ... in the question bank) corrupt shared control-plane /
# node state. A session's setup only touches the questions it actually contains, so damage
# from a task in an earlier session lingers if that task isn't picked again - e.g. Q39 wrecks
# W1's kubelet config, and a later session with different questions never undoes it. This
# restores every known "breaks" target to a good baseline before each session seeds, so
# leftover damage never leaks across sessions regardless of which questions get sampled.
RESTORE_BASELINE = r"""
# apiserver FIRST: Q66 renames the command to "kube-apiserverr", which takes the whole API
# down - every kubectl-based step below would fail until this is undone. The repair goes over
# podman/ssh (nx), not kubectl, so it works precisely when the cluster is unreachable.
if nx "$CP" 'grep -q "^ *- kube-apiserverr$" /etc/kubernetes/manifests/kube-apiserver.yaml' </dev/null 2>/dev/null; then
nx "$CP" 'sed -i "s|^\( *\)- kube-apiserverr$|\1- kube-apiserver|" /etc/kubernetes/manifests/kube-apiserver.yaml' </dev/null
# the kubelet re-reads the manifest on its own; wait for the API to answer again
for _ in $(seq 1 30); do kubectl --context "${CKA_CONTEXT:-kind-cka-sim}" get --raw=/readyz >/dev/null 2>&1 && break; sleep 2; done
fi
# scheduler: undo the "kube-scheduler" -> "kube-schedulerr" command typo (Q20/Q61), and put the
# manifest back if a task had the candidate stop the scheduler by moving it out (Q97) and they
# never moved it back - that setup keeps a pristine copy outside the static pod directory.
nx "$CP" '
f=/etc/kubernetes/manifests/kube-scheduler.yaml
[ -f $f ] || { [ -f /etc/kubernetes/.cka-sim/kube-scheduler.yaml ] && cp /etc/kubernetes/.cka-sim/kube-scheduler.yaml $f; }
sed -i "s|^\( *\)- kube-schedulerr$|\1- kube-scheduler|" $f 2>/dev/null
' </dev/null
# controller-manager: undo Q38's injected flag and Q61's broken image tag. The last two lines
# undo the retired set B's Q317 (the CSR signing controller switched off in --controllers, or in
# its older form --cluster-signing-cert-file pointed at a missing file, which crash-looped the WHOLE
# controller-manager); a cluster that ever ran that set may still carry either.
nx "$CP" '
f=/etc/kubernetes/manifests/kube-controller-manager.yaml
sed -i "/--leader-elect-lease-duration=1s/d" $f 2>/dev/null
sed -i "s|\(image: registry.k8s.io/kube-controller-manager:v[0-9.]*\)-broken|\1|" $f 2>/dev/null
sed -i "s|--cluster-signing-cert-file=/etc/kubernetes/pki/ca-signing.crt|--cluster-signing-cert-file=/etc/kubernetes/pki/ca.crt|" $f 2>/dev/null
sed -i "s|,-certificatesigningrequest-signing-controller||; s|,-csrsigning||" $f 2>/dev/null
' </dev/null
# coredns: restore the default upstream forward (Q37/Q57 point it somewhere bogus)
cf=$($K -n kube-system get cm coredns -o jsonpath='{.data.Corefile}' 2>/dev/null)
if [ -n "$cf" ] && ! printf '%s' "$cf" | grep -Eq '^[[:space:]]*forward \. /etc/resolv\.conf'; then
printf '%s' "$cf" | sed -E 's|^([[:space:]]*forward \. )[^[:space:]{]+|\1/etc/resolv.conf|' > /tmp/Corefile.restore
$K -n kube-system create cm coredns --from-file=Corefile=/tmp/Corefile.restore --dry-run=client -o yaml | $K apply -f - >/dev/null 2>&1
$K -n kube-system rollout restart deploy coredns >/dev/null 2>&1
fi
# ...and Q57's other two faults. kindnet ENFORCES NetworkPolicy, so a leftover lockdown policy on
# the CoreDNS pods kills name resolution for the whole cluster in every later session, and so does
# a Deployment left at 0 replicas.
$K -n kube-system delete netpol maintenance-lockdown --ignore-not-found >/dev/null 2>&1
r=$($K -n kube-system get deploy coredns -o jsonpath='{.spec.replicas}' 2>/dev/null)
if [ -n "$r" ] && [ "$r" -lt 2 ]; then $K -n kube-system scale deploy coredns --replicas=2 >/dev/null 2>&1; fi
# w1/w2/w3 kubelet: undo Q17/Q39-style tampering, then make sure it's enabled and running.
# staticPodPath is normalised unconditionally: Q46 relocates it to /etc/kubernetes/static, and
# the config.yaml.orig copy below only helps if something made that backup first - nothing does,
# so a relocated path used to survive every later session and silently break Q05 (whose check
# looks in /etc/kubernetes/manifests). Only restart the kubelet if we actually changed the file.
# clientCAFile gets the same treatment: Q39 points it at ca-bundle.crt and Q60 at ca-wrong.crt, and
# only W1 ever gets a config.yaml.orig (Q39 makes it), so an unfinished Q60 left W2's kubelet
# crash-looping in every later session - enable/start below cannot help a kubelet that exits.
# The systemd drop-in is checked too: Q92 points ExecStart at a kubelet binary that does not exist.
for n in "$W1" "$W2" "$W3"; do
nx "$n" '
c=/var/lib/kubelet/config.yaml; d=/etc/systemd/system/kubelet.service.d/10-kubeadm.conf
[ -f $c.orig ] && cp -f $c.orig $c
restart=0
if grep -q "^staticPodPath:" $c 2>/dev/null && ! grep -q "^staticPodPath: /etc/kubernetes/manifests$" $c; then
sed -i "s|^staticPodPath: .*|staticPodPath: /etc/kubernetes/manifests|" $c; restart=1
fi
ca=$(awk "/clientCAFile:/{print \$2}" $c 2>/dev/null)
if [ -n "$ca" ] && [ ! -f "$ca" ]; then
sed -i "s|clientCAFile: .*|clientCAFile: /etc/kubernetes/pki/ca.crt|" $c; restart=1
fi
bin=$(sed -n "s|^ExecStart=\([^ ]*kubelet\) .*|\1|p" $d 2>/dev/null | tail -1)
if [ -n "$bin" ] && [ ! -x "$bin" ]; then
sed -i "s|^ExecStart=$bin |ExecStart=/usr/bin/kubelet |" $d; systemctl daemon-reload; restart=1
fi
[ $restart = 1 ] && systemctl restart kubelet >/dev/null 2>&1
systemctl enable kubelet >/dev/null 2>&1
systemctl start kubelet >/dev/null 2>&1
' </dev/null
done
# static pods the bank drops on nodes: Q05's static-web on W3, Q46's relocated copies, Q94's
# ops-beacon on the control plane - and edge-cache, left on W3 by the retired set B (it sat in
# ContainerCreating for days, its hostPath deliberately missing). Each setup recreates what its
# own session needs.
nx "$W3" 'rm -f /etc/kubernetes/manifests/static-web*.yaml /etc/kubernetes/manifests/edge-cache.yaml /etc/kubernetes/static/*.yaml' </dev/null
nx "$CP" 'rm -f /etc/kubernetes/manifests/ops-beacon*.yaml' </dev/null
# default StorageClass: Q16 makes fast-local the default and strips `standard`. Leave exactly one
# default - `standard` - so "the default class" means the same thing in every session.
if $K get sc standard >/dev/null 2>&1; then
for sc in $($K get sc -o jsonpath='{.items[*].metadata.name}'); do
[ "$sc" = standard ] && continue
[ "$($K get sc "$sc" -o jsonpath='{.metadata.annotations.storageclass\.kubernetes\.io/is-default-class}')" = true ] \
&& $K annotate sc "$sc" storageclass.kubernetes.io/is-default-class- >/dev/null 2>&1
done
$K annotate sc standard storageclass.kubernetes.io/is-default-class=true --overwrite >/dev/null 2>&1
fi
# w1-taint: clear every randomized demo taint key the bank can apply. Q25 draws from
# dedicated/team/workload and Q50 from pool/lane; `maintenance` was the retired set B's and may
# still be on a cluster that ran it. Miss one and that node silently stops accepting pods.
# One key per call: `kubectl taint node N a- b-` aborts on the first key that is not present and
# removes NOTHING, so the single taint that is actually set survives the "cleanup" and W1 keeps
# rejecting pods in every later session - the exact failure this block exists to prevent.
for _t in dedicated workload team pool lane maintenance; do
$K taint node "$W1" "$_t-" >/dev/null 2>&1 || true
done
# metrics-server: Q23 seeds its scenario by stripping --kubelet-insecure-tls, and kind's kubelet
# serving certs have no IP SANs, so without that flag metrics-server never becomes Ready. Left
# broken it does far more than disable `kubectl top`: the v1beta1.metrics.k8s.io APIService has no
# endpoints, aggregated discovery goes stale, and then EVERY namespace deletion in the cluster
# hangs on NamespaceDeletionDiscoveryFailure. Baseline runs before seeding, so a session that
# actually contains Q23 re-breaks it a moment later, exactly as intended.
if $K -n kube-system get deploy metrics-server >/dev/null 2>&1 \
&& ! $K -n kube-system get deploy metrics-server -o jsonpath='{.spec.template.spec.containers[0].args}' | grep -q 'kubelet-insecure-tls'; then
$K -n kube-system patch deploy metrics-server --type=json \
-p '[{"op":"add","path":"/spec/template/spec/containers/0/args/-","value":"--kubelet-insecure-tls"}]' >/dev/null 2>&1 || true
fi
# cordons: Q60 cordons W2, and a drain task left half-done leaves a node cordoned. A cordon left
# behind fails any later task that needs every worker (Q26's DaemonSet coverage, Q50's spread).
for n in "$W1" "$W2" "$W3"; do $K uncordon "$n" >/dev/null 2>&1 || true; done
# Deliberately NOT restored, and why:
# etcd (Q45) and certs (Q65). Neither setup damages anything - Q45 only takes a snapshot
# and deletes a ConfigMap, Q65 only records the current notBefore. Any damage comes from a
# half-finished candidate attempt (etcd.yaml left pointing at a restored data-dir, static pod
# manifests moved to /tmp and never moved back), which is not deterministic: there is no way
# to tell "mid-attempt" from "correctly completed" without guessing, and guessing wrong here
# rewrites etcd's data-dir or re-issues certs on a healthy cluster. Re-run ./setup.sh (or
# ./teardown.sh && ./setup.sh) if one of those leaves the control plane wedged.
"""
def restore_baseline(log: Callable[..., Any] = _log_msg):
log(dim(" resetting control-plane / node state to a clean baseline..."))
rc, out = run(RESTORE_BASELINE, timeout=90)
if rc != 0:
log(yellow(f" warning: baseline restore had issues (continuing anyway):\n{dim(out[-500:])}"))
def seed(qs, log: Callable[..., Any] = _log_msg, params_by_id=None):
params_by_id = params_by_id or {}
normal = [q for q in qs if not q.get("disruptive")]
disruptive = [q for q in qs if q.get("disruptive")]
def one(q):
try:
rc, out = run(q["setup"], params=params_by_id.get(str(q["id"])), timeout=240)
except subprocess.TimeoutExpired:
rc, out = 1, "setup timed out after 240s"
log(f" {'✓' if rc == 0 else '!'} Q{q['id']:02d} {q['title']}" + ("" if rc == 0 else f"\n{dim(out[-300:])}"))
for q in normal:
one(q)
if disruptive:
log(dim(" waiting up to 45s for seeded pods to be scheduled..."))
for _ in range(15):
pending = subprocess.run(["kubectl", "--context", CONTEXT, "get", "pods", "-A", "--field-selector=spec.nodeName=",
"-o", "name"], capture_output=True, text=True).stdout.strip()
if not pending:
break
time.sleep(3)
for q in disruptive:
one(q)
def begin_session(qs, mode, duration, force, log: Callable[..., Any] = _log_msg, hard=False, untimed_session=False):
st = load_state()
if st and not st.get("ended_at") and not force:
die("A session is already in progress. './exam end' it first, or add --force.")
if not kubectl_ok():
die(f"Cannot reach the cluster through kubectl context '{CONTEXT}'. Run ./setup.sh (kind) or set CKA_CONTEXT for another cluster.")
if not qs:
die(f"Track '{TRACK_ID}' has no questions yet.")
preflight()
os.makedirs(WORKSPACE, exist_ok=True)
rng = random.Random()
params_by_id = {str(q["id"]): make_params(q, rng) for q in qs}
setup_contexts(hard, params_by_id)
restore_baseline(log)
log(bold(f"Seeding {len(qs)} scenario(s){' [HARD]' if hard else ''}..."))
seed(qs, log, params_by_id)
st = {"mode": mode, "track": TRACK_ID, "hard": hard, "untimed": untimed_session, "ids": [q["id"] for q in qs],
"params": params_by_id, "duration": duration, "started_at": time.time(), "ended_at": None,
"results": {}, "flags": [], "notes": ""}
save_state(st)
log("")
log(bold("=" * 64))
log(bold(f" {TRACK_ID.upper()} {'EXAM' if mode == 'exam' else 'DRILL'}{' · HARD' if hard else ''} - clock started"))
if hard:
log(yellow(" Contexts: k8s-c1-H (default) · k8s-c2-AC · k8s-c3-CCC - each task tells you which to use; you are on k8s-c1-H now"))
log(f" Ends : {datetime.fromtimestamp(st['started_at'] + duration).strftime('%H:%M:%S')} ({duration // 60} min)")
log(f" Tasks : {', '.join('Q%02d' % q['id'] for q in qs) if mode == 'drill' else len(qs)} Pass mark: {PASS_PCT}%")
log(bold("=" * 64))
log(" ./exam web exam-style browser UI (task pane + terminal + clock)\n"
" ./exam list / show N / grade N / timer / end terminal workflow\n"
+ (" Tip: read ALL tasks first. Some are cluster-wide failures that block others.\n" if mode == "exam" else "")
+ f" Files the tasks ask for go under {WS_DISPLAY}/ (relative to this directory)\n"
+ " Your own terminal: source ./examrc (fish: source ./examrc.fish) -> k, $do, $now, completion")
return st
def finish_session(st, log: Callable[..., Any] = _log_msg):
qs = session_questions(st)
over = remaining(st) <= 0
set_paused(st, False)
log(bold(f"Submitting - grading {len(qs)} task(s)..."))
for q in qs:
r = grade_one(q)
st["results"][str(q["id"])] = r
tag = green("PASS") if r["pass"] else (yellow(f"{r['earned']}/{r['total']}") if r["earned"] else red("FAIL"))
log(f"Q{q['id']:02d} {tag} {q['title']}" + ("" if r["pass"] else f"\n{dim(result_detail(r))}"))
st["ended_at"] = time.time()
st["over_time"] = over
save_state(st)
setup_contexts(False, {})
if untimed(st):
return None
return record_attempt(st)
# ----------------------------------------------------------- commands ----
def cmd_start(args):
hard = "--hard" in args; force = "--force" in args
undrilled = [d for d, n in drilled_domains().items() if n == 0]
if undrilled and not force:
die("Drill every domain at least once before a full exam (that is the point of the drills).\n"
f" Not yet drilled: {', '.join(undrilled)}\n Run: ./exam drill (picks the next domain)\n"
" or: ./exam start --force to skip the gate.")
begin_session(compose_exam(hard, random.Random()), "exam", EXAM_SECONDS, force, hard=hard)
def next_drill(hard, rng):
"""No-arg drill: the least-drilled domain, DRILL_SIZE tasks, least-recently-seen first."""
cov = drilled_domains()
domain = min(DOMAINS, key=lambda d: (cov[d], list(DOMAINS).index(d)))
return domain, pick(domain, DRILL_SIZE, hard, rng)
def cmd_drill(args):
force = "--force" in args; hard = "--hard" in args
args = [a for a in args if a not in ("--force", "--hard")]
minutes = None
if "--time" in args:
i = args.index("--time"); minutes = int(args[i + 1]); del args[i:i + 2]
per_task = DRILL_SECONDS_PER_TASK_HARD if hard else DRILL_SECONDS_PER_TASK
if not args:
domain, qs = next_drill(hard, random.Random())
cov = drilled_domains()
print(bold(f"Drill: {DOMAINS[domain]}") + dim(f" ({sum(1 for v in cov.values() if v)}/{len(DOMAINS)} domains drilled so far)"))
begin_session(qs, "drill", (minutes * 60) if minutes else per_task * len(qs), force, hard=hard)
return
if args[0] == "all":
label, qs = campaign_batch(hard, random.Random())
if not qs:
print(bold("Every task in the bank has been attempted." ) + " Switching to weak tasks.")
qs, _ = split_solo([q for q in QUESTIONS if q["id"] in weak_ids()])
qs = no_conflicts(qs, DRILL_SIZE)
if not qs:
die("Nothing weak either - you have passed everything on its latest attempt. Time for ./exam start --hard.")
else:
print(bold(f"Campaign: {label}") + dim(f" this batch: Q{', Q'.join(str(q['id']) for q in qs)}"))
begin_session(qs, "drill", (minutes * 60) if minutes else per_task * len(qs), force, hard=hard)
return
if args[0] == "weak":
ids = weak_ids()
if not ids:
die("Nothing marked weak - finish a session with './exam end' first, or name tasks: ./exam drill 2 17")
else:
ids = []
for a in args:
if a.lower().replace("set", "") in SETS and a.lower().startswith("set"):
ids += [q["id"] for q in set_questions(a.lower().replace("set", ""))]
elif a.upper() in DOMAINS:
ids += [q["id"] for q in QUESTIONS if q["domain"] == a.upper()]
elif a.isdigit():
ids.append(q_by_id(int(a))["id"])
else:
die(f"'{a}' is not a task id, a domain ({', '.join(DOMAINS)}) or a set "
f"({', '.join('set' + k for k in SETS) or 'none'}).")
ids = sorted(set(ids))
qs, solo = split_solo([q for q in QUESTIONS if q["id"] in ids])
if solo:
print(yellow(f"Dropped Q{', Q'.join(str(q['id']) for q in solo)}: restoring it rolls the whole cluster back and would "
f"wipe the other tasks. Drill it on its own: ./exam drill {solo[0]['id']}"))
kept = no_conflicts(qs, len(qs))
if len(kept) < len(qs):
dropped = [q["id"] for q in qs if q not in kept]
print(yellow(f"Dropped Q{', Q'.join(map(str, dropped))}: they break the same thing as another task in this set. Drill them separately."))
# Shuffle AFTER resolving conflicts, so which task gets dropped stays deterministic while the
# order you meet them in does not. A drill listed by id walks you from easy to hard in
# lockstep, which is the one ordering the real exam will never give you.
random.Random().shuffle(kept)
begin_session(kept, "drill", (minutes * 60) if minutes else per_task * len(kept), force, hard=hard)
def cmd_study(args):
"""No clock, solutions unlocked from the start, nothing written to history."""
hard = "--hard" in args; force = "--force" in args
rest = [a for a in args if not a.startswith("--")]
n = DRILL_SIZE
if "--count" in args:
n = int(args[args.index("--count") + 1]); rest = [a for a in rest if a != str(n)]
rng = random.Random()
if rest and rest[0].upper() in DOMAINS:
domain = rest[0].upper()
qs = pick(domain, n, hard, rng)
print(bold(f"Study: {DOMAINS[domain]}"))
elif rest and rest[0] == "weak":
qs = no_conflicts(split_solo([q for q in QUESTIONS if q["id"] in weak_ids()])[0], n)
print(bold("Study: tasks you have been failing"))
elif rest:
qs = no_conflicts(split_solo([q_by_id(int(a)) for a in rest])[0], len(rest))
print(bold("Study: " + ", ".join(f"Q{q['id']}" for q in qs)))
else:
domain, qs = next_drill(hard, rng)
print(bold(f"Study: {DOMAINS[domain]}"))
begin_session(qs, "drill", EXAM_SECONDS, force, hard=hard, untimed_session=True)
def cmd_sim(args):
"""A fixed simulator set, full exam rules and clock."""
force = "--force" in args
names = [a.lower() for a in args if not a.startswith("--")]
if not names or names[0] not in SETS:
die(f"usage: ./exam sim {'|'.join(SETS) or '<set>'} [--force]\n "
+ "\n ".join(f"{k}: {v} ({len(set_questions(k))} tasks)" for k, v in SETS.items()))
name = names[0]
qs = set_questions(name)
print(bold(f"{SETS[name]} - {len(qs)} tasks, {TRACK['minutes']} minutes, exam rules (no feedback until you submit)"))
begin_session(qs, "exam", EXAM_SECONDS, force, hard=True)
def cmd_list(args):
st = load_state()
results = st["results"] if st and "results" in st else {}
qs = session_questions(st) if st and not st.get("ended_at") else list(QUESTIONS)
agg = stats()
print(bold(f"{'#':>3} {'Domain':<8}{'Pts':>4} {'Now':<8} {'History':<9} Title"))
for q in qs:
r = results.get(str(q["id"]))
now = "-" if r is None else (green("PASS") if r.get("pass") else yellow(f"{r.get('earned')}/{r.get('total')}"))
a = agg.get(int(q["id"])) or AggEntry()
hist = "" if a.attempts == 0 else f"{a.attempts - a.fails}/{a.attempts}"
if a.attempts and a.last is False:
hist = red(hist + " ✗")
print(f"{q['id']:>3} {q['domain']:<8}{q['weight']:>4} {now:<17} {hist:<18} {q['title']}{' [hard]' if q['difficulty'] == 'hard' else ''}")
print(dim("\nHistory = full passes/attempts across sessions, ✗ = failed most recently."))
if st and not st.get("ended_at"):
print(status_line(st))
def cmd_show(args):
if not args:
die("usage: ./exam show N")
q = q_by_id(int(args[0]))
st0 = load_state()
sq = session_questions(st0) if st0 and not st0.get("ended_at") else []
pos = f"{sq.index(q) + 1} of {len(sq)} | " if q in sq else ""
print(bold(f"{pos}Q{q['id']} | {DOMAINS[q['domain']]} | {q['weight']} pts{' | HARD' if q['difficulty'] == 'hard' else ''}"))
print(bold(q["title"])); print("-" * 64); print(task_text(q)); print("-" * 64)
st = load_state()
if st and not st.get("ended_at"):
print(status_line(st))
def cmd_grade(args):
st = require_started()
force = "--force" in args; args = [a for a in args if a != "--force"]
if not args:
die("usage: ./exam grade N | all")
# The web UI already refused this; the CLI handed out per-check feedback mid-exam, which a sim
# meant to rehearse the real thing (no feedback until you submit) must not do by default.
if st.get("mode") == "exam" and not st.get("ended_at") and not force:
die("No feedback during an exam - the real one gives none. Submit with ./exam end (--force to peek anyway).")
ids = [q["id"] for q in session_questions(st)] if args[0] == "all" else [int(a) for a in args]
for n in ids:
q = q_by_id(n); r = grade_one(q)
st["results"][str(n)] = r
tag = green("PASS") if r.get("pass") else yellow(f"{r.get('earned')}/{r.get('total')}")
print(f"Q{n:02d} {tag} {q['title']}"); print(dim(result_detail(r)))
save_state(st)
if args[0] == "all":
print_score(st)
def print_score(st):
qs = session_questions(st)
earned, total, pct = score(st)
print(); print(bold("Per-domain:"))
for d, name in DOMAINS.items():
dq = [q for q in qs if q["domain"] == d]
if not dq:
continue
dt = sum(q["weight"] for q in dq); de = sum(st["results"].get(str(q["id"]), {}).get("earned", 0) for q in dq)
n = int(20 * de / dt)
print(f" {d:<7} {'█' * n}{'░' * (20 - n)} {de:>3}/{dt:<3} {dim(name)}")
verdict = green("PASS") if pct >= PASS_PCT else red("FAIL")
print(f"\n{bold('Score:')} {earned}/{total} = {pct}% -> {verdict} (pass mark {PASS_PCT}%)")
def status_line(st):
if untimed(st):
return green("study mode - no clock") + dim(f" [{len(session_questions(st))} tasks, nothing recorded]")
r = remaining(st)
if st.get("paused_at"):
return yellow(f"⏸ PAUSED - {fmt(r)} remaining") + dim(" ./exam resume")
if r <= 0:
return red("⏰ TIME IS UP - run ./exam end")
col = green if r > 30 * 60 else (yellow if r > 10 * 60 else red)
return col(f"⏱ {fmt(r)} remaining") + dim(f" [{st.get('mode', 'exam')}]")
def cmd_status(args):
st = require_started()
if st.get("ended_at"):
print("Session ended."); print_score(st); return
print(status_line(st))
print(dim(f"{sum(1 for v in st['results'].values() if v['pass'])} passed / {len(st['results'])} graded / {len(session_questions(st))} in session"))
def cmd_timer(args):
st = require_started()
if st.get("ended_at"):
die("Session already ended.")
if untimed(st):
die("This is a study session - there is no clock.")
try:
while True:
st = load_state(); r = remaining(st)
if st.get("paused_at"):
sys.stdout.write("\r" + yellow(bold(f"⏸ {fmt(r)} PAUSED ")) + dim(" ./exam resume ")); sys.stdout.flush(); time.sleep(1); continue
if r <= 0:
print("\r" + red(bold("⏰ 00:00:00 TIME IS UP - run ./exam end "))); break
col = green if r > 30 * 60 else (yellow if r > 10 * 60 else red)
sys.stdout.write("\r" + col(bold(f"⏱ {fmt(r)} remaining")) + dim(" (Ctrl-C to hide)")); sys.stdout.flush()
if int(r) in (30 * 60, 15 * 60, 5 * 60) and TTY:
sys.stdout.write("\a")
time.sleep(1)
except KeyboardInterrupt:
print()
def cmd_pause(args):
st = require_started()
if st.get("ended_at"):
die("Session already ended.")
set_paused(st, True); print(status_line(st))
def cmd_resume(args):
st = require_started()
set_paused(st, False); print(status_line(st))
def cmd_end(args):
st = require_started()
if st.get("ended_at"):
die("Session already ended and recorded.")
if remaining(st) <= 0:
print(yellow("Note: time expired before submission."))
entry = finish_session(st)
print_score(st)
if entry is None:
print(dim("Study session - not added to history or weak-spots.")); return
print(dim(f"Time used: {fmt(st['ended_at'] - st['started_at'] - paused_total(st))}" + (" (over time)" if st.get("over_time") else "")))
failed = sorted(int(k) for k, v in entry["results"].items() if not v["pass"])
print(dim(f"Attempt #{len(load_history())} logged to {os.path.basename(HISTORY_FILE)}; {os.path.basename(WEAK_FILE)} updated."))
if failed:
print(bold("\nTo drill what you missed: ./exam drill weak") + dim(f" (Q{', Q'.join(map(str, failed))})"))
def cmd_abort(args):
"""End a session without grading it and restore the kubectl context."""
st = load_state()
if not st or st.get("ended_at"):
die("No session in progress.")
assert st is not None
if "--yes" not in args:
die(f"This discards the {st.get('mode', 'exam')} in progress without grading or recording it.\n"
" Confirm: ./exam abort --yes (to grade and keep it instead: ./exam end)")
st["ended_at"] = time.time(); st["aborted"] = True
save_state(st)
setup_contexts(False, {})
print(yellow("Session discarded. Nothing was graded or written to history."))
print(dim(f" kubectl context restored to {CONTEXT}. Seeded objects are still on the cluster;"))
print(dim(" './exam drill ...' or './exam start' re-seeds, './teardown.sh' removes the cluster."))
def cmd_weak(args):
agg = stats(); hist = load_history()
if not hist:
die("No history yet - finish a session with './exam end'.")
ids = [i for i in weak_ids(agg) if agg[i]["attempts"] > 0]
print(bold(f"Weak spots after {len(hist)} attempt(s):"))
if not ids:
print(green(" none - everything attempted has passed on its latest try"))
for i in ids:
q = q_by_id(i); a = agg[i]
print(f"\n{red('Q%02d' % i)} {q['domain']:<7} {q['title']} " + dim(f"{a['fails']}/{a['attempts']} failed, last {a['last_earned']} pts, last failed {a['last_fail_at']}"))
for n in weak_lines(a):
print(dim(" " + n))
never = [q["id"] for q in QUESTIONS if agg[q["id"]]["attempts"] == 0]
if never:
print(dim(f"\nNever attempted: Q{', Q'.join(map(str, never))}"))
print(dim(f"\nFull log: {WEAK_FILE} Drill them: ./exam drill weak"))
def cmd_history(args):
hist = load_history(include_synthetic=True)
if not hist:
die("No history yet.")
print(bold(f"{'#':>3} {'when':<17}{'mode':<7}{'score':>7} {'time':<9} tasks"))
for i, e in enumerate(hist, 1):
col = green if e["score_pct"] >= PASS_PCT else red
tasks = f"all {len(QUESTIONS)}" if len(e["ids"]) == len(QUESTIONS) else "Q" + ",".join(map(str, e["ids"]))
tag = dim(" [scripted verification - not counted]") if e.get("synthetic") else ""
print(f"{i:>3} {e['at']:<17}{(e['mode'] + ('*' if e.get('hard') else '')):<7}{col('%5.1f%%' % e['score_pct']):>16} {fmt(e['time_used']):<9} {tasks}{tag}")
exams = [e for e in hist if e["mode"] == "exam" and not e.get("synthetic")]
print(dim("* = hard"))
if len(exams) >= 2:
print(bold("\nDomain trend (full exams, % of that exam's points in the domain):"))
for d in DOMAINS:
cells = []
for e in exams:
de, dt = exam_domain_scores(e).get(d, (0, 0))
cells.append(f"{round(100 * de / dt):>3}%" if dt else " -")
print(f" {d:<7} " + " → ".join(cells))
def cmd_reset(args):
if not args:
die("usage: ./exam reset N (re-seed one task's scenario)")
q = q_by_id(int(args[0])); rc, out = run(str(q.get("setup", "")), params=session_params(load_state(), q))
print(("re-seeded " if rc == 0 else "problem re-seeding ") + f"Q{q['id']:02d}: {q['title']}")
if out:
print(dim(out))
# --------------------------------------------------------------- web ----
class Api(http.server.BaseHTTPRequestHandler):
"""Serves web/index.html plus a small JSON API. State lives in the same files the CLI uses."""
lock = threading.Lock()
log_lines = []
seeding = False
seed_error = None
ttyd_port = 7681
shell = "bash"
def log_message(self, format: str, *args: Any) -> None: # quiet
pass
def _json(self, obj, code=200):
body = json.dumps(obj).encode()
self.send_response(code); self.send_header("Content-Type", "application/json")
self.send_header("Content-Length", str(len(body))); self.end_headers(); self.wfile.write(body)
def _read_json(self):
n = int(self.headers.get("Content-Length") or 0)
return json.loads(self.rfile.read(n) or b"{}")
def do_GET(self):
p = urllib.parse.urlparse(self.path)
if p.path in ("/", "/index.html"):
with open(os.path.join(ROOT, "web", "index.html"), "rb") as f:
body = f.read()
self.send_response(200); self.send_header("Content-Type", "text/html; charset=utf-8")
self.send_header("Content-Length", str(len(body))); self.end_headers(); self.wfile.write(body); return
if p.path == "/api/session":
with Api.lock:
return self._json(self.session_payload())
if p.path == "/api/history":
return self._json({"history": load_history(), "weak": weak_ids(), "stats": {str(k): v.to_dict() for k, v in stats().items()}})
self._json({"error": "not found"}, 404)
def do_POST(self):
p = urllib.parse.urlparse(self.path); body = self._read_json()
with Api.lock:
st = load_state()
if p.path == "/api/track":
t = body.get("track")
if t not in TRACKS:
return self._json({"error": f"unknown track {t}"}, 400)
if Api.seeding:
return self._json({"error": "a session is being seeded"}, 409)
load_track(t)
Api.log_lines = []
return self._json(self.session_payload())
if p.path == "/api/start":
if Api.seeding:
return self._json({"error": "already seeding"}, 409)
mode = body.get("mode", "exam"); hard = bool(body.get("hard")); rng = random.Random()
Api.log_lines = []; Api.seed_error = None; Api.seeding = True
ansi = __import__("re").compile(r"\x1b\[[0-9;]*m")
ulog = lambda line: Api.log_lines.append(ansi.sub("", line))
def work():
try:
if body.get("sim") in SETS:
begin_session(set_questions(body["sim"]), "exam", EXAM_SECONDS, True, log=ulog, hard=True)
elif mode == "exam":
begin_session(compose_exam(hard, rng), "exam", EXAM_SECONDS, True, log=ulog, hard=hard)
else:
if body.get("domain") in DOMAINS:
qs = pick(body["domain"], DRILL_SIZE, hard, rng)
elif body.get("campaign"):
_, qs = campaign_batch(hard, rng)
if not qs:
qs = no_conflicts(split_solo([q for q in QUESTIONS if q["id"] in weak_ids()])[0], DRILL_SIZE)
elif body.get("next"):
_, qs = next_drill(hard, rng)
else:
ids = body.get("ids") or weak_ids()
qs = no_conflicts(split_solo([q for q in QUESTIONS if q["id"] in ids])[0], len(ids))
per = DRILL_SECONDS_PER_TASK_HARD if hard else DRILL_SECONDS_PER_TASK
begin_session(qs, "drill", per * len(qs), True, log=ulog, hard=hard,
untimed_session=bool(body.get("study")))
except SystemExit:
Api.seed_error = "\n".join(l for l in Api.log_lines if l.strip()) or "start failed - see the terminal running ./exam web"
except Exception as e: # noqa: BLE001
Api.seed_error = f"{type(e).__name__}: {e}"
finally:
Api.seeding = False
threading.Thread(target=work, daemon=True).start()
return self._json({"seeding": True})
if not st:
return self._json({"error": "no session"}, 400)
if p.path == "/api/flag":
qid = int(body["id"]); flags = set(st.get("flags", []))
flags.symmetric_difference_update({qid}); st["flags"] = sorted(flags); save_state(st)
elif p.path == "/api/pause":
set_paused(st, bool(body.get("paused", True)))
elif p.path == "/api/notes":
st["notes"] = body.get("notes", ""); save_state(st)
elif p.path == "/api/grade":
if st.get("mode") != "drill" and not st.get("ended_at"):
return self._json({"error": "No feedback during an exam. Submit to see results."}, 403)
q = q_by_id(int(body["id"]))
if "results" not in st or not isinstance(st["results"], dict):
st["results"] = {}
st["results"][str(q["id"])] = grade_one(q, session_params(st, q)); save_state(st)
elif p.path == "/api/end":
if not st.get("ended_at"):
finish_session(st, log=lambda *_: None)
elif p.path == "/api/reset":
q = q_by_id(int(body["id"])); run(str(q.get("setup", "")), params=session_params(st, q))
else:
return self._json({"error": "not found"}, 404)
return self._json(self.session_payload())
def session_payload(self) -> dict[str, Any]:
st = load_state()
qs = session_questions(st) if st else QUESTIONS
payload: dict[str, Any] = {
"track": TRACK, "track_id": TRACK_ID, "workspace": WS_DISPLAY, "version": VERSION, "ttyd_port": Api.ttyd_port,
"ttyd": shutil.which("ttyd") is not None, "shell": Api.shell, "setup_log": Api.log_lines,
"questions": [{"id": q["id"], "domain": q["domain"], "title": q["title"], "weight": q["weight"], "difficulty": q["difficulty"],
"task": task_text(q, session_params(st, q)), "checks": [{"name": c["name"], "pts": c["pts"]} for c in q["checks"]],
"solution": render(str(q.get("solution", "")), session_params(st, q)) if solution_allowed(st, int(q["id"])) else None}
for q in qs],
"bank": [{"id": q["id"], "domain": q["domain"], "title": q["title"], "difficulty": q["difficulty"]} for q in QUESTIONS],
"tracks": TRACKS, "drilled": drilled_domains(), "sets": {k: {"name": v, "count": len(set_questions(k))} for k, v in SETS.items()},
"seen": sorted(last_seen()),
"seeding": Api.seeding, "seed_error": Api.seed_error,
"session": None,
}
if st:
earned, total, pct = score(st)
payload["session"] = {"mode": st.get("mode"), "hard": st.get("hard", False), "started_at": st["started_at"], "duration": st.get("duration", EXAM_SECONDS),
"ends_at": st["started_at"] + st.get("duration", EXAM_SECONDS), "ended_at": st.get("ended_at"),
"over_time": st.get("over_time", False), "flags": st.get("flags", []),
"paused": bool(st.get("paused_at")), "paused_total": paused_total(st),
"untimed": untimed(st), "notes": st.get("notes", ""),
"results": st["results"], "score": {"earned": earned, "total": total, "pct": pct, "pass_pct": PASS_PCT},
"domains": self.domain_scores(st), "server_time": time.time()}
return payload
def domain_scores(self, st: dict[str, Any]) -> dict[str, Any]:
qs = session_questions(st); out = {}
st_results: dict[str, Any] = st.get("results", {})
for d in DOMAINS:
dq = [q for q in qs if q["domain"] == d]
if dq:
out[d] = {"earned": sum(st_results.get(str(q["id"]), {}).get("earned", 0) for q in dq),
"total": sum(q["weight"] for q in dq)}
return out
def cmd_web(args):
port = 8080
if "--port" in args:
port = int(args[args.index("--port") + 1])
Api.shell = os.environ.get("CKA_SHELL", "bash")
if "--shell" in args:
Api.shell = args[args.index("--shell") + 1]
os.makedirs(WORKSPACE, exist_ok=True)
import socket
def port_free(pt):
with socket.socket() as sk:
sk.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
try:
sk.bind(("127.0.0.1", pt)); return True
except OSError:
return False
if not port_free(port):
die(f"Port {port} is already in use - probably another './exam web' still running.\n"
f" Find it: ss -ltnp | grep :{port} or start on another port: ./exam web --port {port + 10}")
if shutil.which("ttyd") and not port_free(Api.ttyd_port):
print(yellow(f" ttyd port {Api.ttyd_port} already in use - reusing the existing terminal server."))
ttyd = None
if shutil.which("ttyd") and port_free(Api.ttyd_port):
env = dict(os.environ, WORKSPACE=WORKSPACE, KUBECONFIG=os.environ.get("KUBECONFIG", os.path.expanduser("~/.kube/config")))
rc = os.path.join(ROOT, "examrc.fish" if "fish" in Api.shell else "examrc")
if "fish" in Api.shell:
shell_cmd = [Api.shell, "-C", f"source {rc}"]
else:
shell_cmd = [Api.shell, "--rcfile", rc, "-i"]
# -W makes the terminal writable, so it must never be reachable off this machine: it is an
# unauthenticated root-equivalent shell with the cluster kubeconfig. ttyd binds every
# interface by default, so pin it to loopback (macOS needs lo0; override if yours differs).
iface = os.environ.get("CKA_TTYD_INTERFACE", "lo")
ttyd = subprocess.Popen(["ttyd", "-p", str(Api.ttyd_port), "-i", iface, "-W",
"-t", "fontSize=14", "-t", "theme={\"background\":\"#1a1b1e\"}",
"--cwd", ROOT] + shell_cmd, env=env,
stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
print(dim(f" terminal: ttyd on {iface}:{Api.ttyd_port} running {Api.shell} in {ROOT} with k=kubectl, $do and $now preset"))
else:
print(yellow(" ttyd not found - the terminal pane will show install instructions.\n"
" (dnf/apt/pacman/brew install ttyd, or https://github.com/tsl0922/ttyd/releases)"))
http.server.ThreadingHTTPServer.allow_reuse_address = True
srv = http.server.ThreadingHTTPServer(("127.0.0.1", port), Api) # type: ignore[arg-type]
print(bold(f"\n {TRACK_ID.upper()} exam UI: http://127.0.0.1:{port} (exam {VERSION})\n") + dim(" Ctrl-C stops the UI (not the exam clock)."))
def stop(*_):
if ttyd:
ttyd.terminate()
srv.shutdown()
signal.signal(signal.SIGINT, lambda *_: threading.Thread(target=stop).start())
signal.signal(signal.SIGTERM, lambda *_: threading.Thread(target=stop).start())
try:
srv.serve_forever()
finally:
if ttyd:
ttyd.terminate()
def cmd_version(args):
print(f"exam {VERSION} track={TRACK_ID} bank={len(QUESTIONS)} tasks python={sys.version.split()[0]}")
def cmd_help(args):
print(__doc__ + f"\nversion {VERSION}")
print(f"""usage: ./exam [--track cka|ckad|cks] <command> (current track: {TRACK_ID}, {len(QUESTIONS)} tasks)
web [--port 8080] [--shell bash] exam-style browser UI: tasks, terminal, clock, submit
study [DOMAIN|weak|N..] no clock, solutions unlocked, nothing recorded (--count N, default {DRILL_SIZE})
drill next domain you have drilled least: {DRILL_SIZE} tasks, 6 min each (--hard: 5 min, hard tasks)
drill all [--hard] campaign: {DRILL_SIZE} never-attempted tasks per session until the bank is done, then weak ones
drill weak|N..|DOMAIN.. chosen tasks (--time MIN, --force, --hard)
progress attempted/passed per domain, normal and hard
solution N reference solution (drills: after grading N; exams: after submit)
sim {'|'.join(SETS) or '<set>'}{' ' * max(1, 20 - len('|'.join(SETS)))}fixed simulator set, exam rules, {TRACK['minutes']} min
start [--hard] full exam: {sum(QUOTA.values()) or '?'} tasks sampled by domain ({sum(QUOTA_HARD.values()) or '?'} in hard mode), {TRACK['minutes']} min.
Requires every domain drilled once first (--force skips). Hard mode adds traps,
exact-output tasks, compound failures, collateral checks and exam-style contexts.
list tasks in this session, with current + historical results
show N display task N
grade N|all grade one task (or all) per check - drills only (--force to peek in an exam)
status / timer time remaining (timer = live countdown)
pause / resume stop the clock (not recorded anywhere - but the real exam has no pause)
end submit: grade per check, score, log attempt, update weak-spots
weak what you keep getting wrong, with the failing checks
history every attempt, scores, and per-domain trend
reset N re-seed the scenario for task N
""")
COMMANDS = {"start": cmd_start, "drill": cmd_drill, "list": cmd_list, "show": cmd_show, "grade": cmd_grade,
"status": cmd_status, "timer": cmd_timer, "pause": cmd_pause, "resume": cmd_resume, "end": cmd_end, "weak": cmd_weak, "history": cmd_history,
"reset": cmd_reset, "web": cmd_web, "abort": cmd_abort, "progress": cmd_progress, "solution": cmd_solution, "sim": cmd_sim, "study": cmd_study, "version": cmd_version, "help": cmd_help}
if __name__ == "__main__":
cmd = ARGV[0] if ARGV else "help"
COMMANDS.get(cmd, cmd_help)(ARGV[1:])
Reference in New Issue View Git Blame Copy Permalink