Release v0.1.16
Dependency Audit / dependency-audit (push) Failing after 1m47s
Deployment Installer / deployment-installer (push) Successful in 6s
Security Audit / security-audit (push) Successful in 11m18s
Developer Meta-package Release / publish-package (push) Successful in 11s

This commit is contained in:
2026-08-05 19:52:32 +02:00
parent 61463a24cb
commit 3ca068f76a
42 changed files with 5433 additions and 189 deletions
@@ -0,0 +1,85 @@
schema_version = 1
name = "govoplan-k8s-acceptance"
mode = "acceptance"
state_directory = "~/.local/share/govoplan/labs/govoplan-k8s-acceptance"
vm_image_directory = "/var/lib/libvirt/images/govoplan-labs"
ssh_user = "govoplan"
ssh_private_key = "~/.ssh/govoplan-lab"
ssh_public_key = "~/.ssh/govoplan-lab.pub"
namespace = "govoplan"
public_host = "govoplan.acceptance.example.org"
s3_host = "s3.govoplan.acceptance.example.org"
ingress_class = "traefik"
module_set = "base"
api_replicas = 2
web_replicas = 2
worker_replicas = 2
db_connection_limit = 100
[network]
prefix_length = 24
gateway = "10.77.10.1"
dns_servers = ["10.77.10.1", "1.1.1.1"]
bridge = "br0"
[image]
url = "https://cloud-images.ubuntu.com/releases/noble/release-20260801/ubuntu-24.04-server-cloudimg-amd64.img"
sha256 = "0533b0655c32e68b31d792ecd6ccfca95abdbc536c4446874fe0513bd4140ffe"
[k3s]
version = "v1.36.1+k3s1"
binary_url = "https://github.com/k3s-io/k3s/releases/download/v1.36.1%2Bk3s1/k3s"
binary_sha256 = "a443db3fe9820cd93617ae67e4386d87c1514c1e96ceb30f4c2791c39065653c"
install_script_url = "https://raw.githubusercontent.com/k3s-io/k3s/v1.36.1%2Bk3s1/install.sh"
install_script_sha256 = "46177d4c99440b4c0311b67233823a8e8a2fc09693f6c89af1a7161e152fbfad"
[release]
manifest_url = "https://git.add-ideas.de/GovOPlaN/govoplan/releases/download/v0.1.15/distribution-manifest.json"
manifest_sha256 = "09ac1ade6ede4958bab0dfb7fd8f99246f4d991846308db1f410b25b46267840"
keyring_url = "https://git.add-ideas.de/GovOPlaN/govoplan/releases/download/v0.1.15/distribution-keyring.json"
keyring_sha256 = "92a9f8e3bac0ef525ad9a063c76faa36233a9070cd4a5db9ee9b3f92323b602f"
channel = "stable"
[[nodes]]
name = "control-1"
role = "control"
address = "10.77.10.21"
hypervisor = "lab-admin@hypervisor-state.example.org"
failure_domain = "rack-c"
mac_address = "52:54:00:68:01:01"
cpus = 2
memory_mib = 4096
disk_gib = 32
[[nodes]]
name = "worker-1"
role = "worker"
address = "10.77.10.22"
hypervisor = "lab-admin@hypervisor-a.example.org"
failure_domain = "rack-a"
mac_address = "52:54:00:68:01:02"
cpus = 2
memory_mib = 4096
disk_gib = 40
[[nodes]]
name = "worker-2"
role = "worker"
address = "10.77.10.23"
hypervisor = "lab-admin@hypervisor-b.example.org"
failure_domain = "rack-b"
mac_address = "52:54:00:68:01:03"
cpus = 2
memory_mib = 4096
disk_gib = 40
[[nodes]]
name = "state-1"
role = "state"
address = "10.77.10.24"
hypervisor = "lab-admin@hypervisor-state.example.org"
failure_domain = "rack-c"
mac_address = "52:54:00:68:01:04"
cpus = 4
memory_mib = 8192
disk_gib = 120
+85
View File
@@ -0,0 +1,85 @@
schema_version = 1
name = "govoplan-k8s-lab"
mode = "rehearsal"
state_directory = "~/.local/share/govoplan/labs/govoplan-k8s-lab"
vm_image_directory = "/var/lib/libvirt/images/govoplan-labs"
ssh_user = "govoplan"
ssh_private_key = "~/.ssh/govoplan-lab"
ssh_public_key = "~/.ssh/govoplan-lab.pub"
namespace = "govoplan"
public_host = "govoplan.lab.test"
s3_host = "s3.govoplan.lab.test"
ingress_class = "traefik"
module_set = "base"
api_replicas = 2
web_replicas = 2
worker_replicas = 2
db_connection_limit = 100
[network]
prefix_length = 24
gateway = "192.168.123.1"
dns_servers = ["192.168.123.1", "1.1.1.1"]
bridge = "virbr-gplab"
[image]
url = "https://cloud-images.ubuntu.com/releases/noble/release-20260801/ubuntu-24.04-server-cloudimg-amd64.img"
sha256 = "0533b0655c32e68b31d792ecd6ccfca95abdbc536c4446874fe0513bd4140ffe"
[k3s]
version = "v1.36.1+k3s1"
binary_url = "https://github.com/k3s-io/k3s/releases/download/v1.36.1%2Bk3s1/k3s"
binary_sha256 = "a443db3fe9820cd93617ae67e4386d87c1514c1e96ceb30f4c2791c39065653c"
install_script_url = "https://raw.githubusercontent.com/k3s-io/k3s/v1.36.1%2Bk3s1/install.sh"
install_script_sha256 = "46177d4c99440b4c0311b67233823a8e8a2fc09693f6c89af1a7161e152fbfad"
[release]
manifest_url = "https://git.add-ideas.de/GovOPlaN/govoplan/releases/download/v0.1.15/distribution-manifest.json"
manifest_sha256 = "09ac1ade6ede4958bab0dfb7fd8f99246f4d991846308db1f410b25b46267840"
keyring_url = "https://git.add-ideas.de/GovOPlaN/govoplan/releases/download/v0.1.15/distribution-keyring.json"
keyring_sha256 = "92a9f8e3bac0ef525ad9a063c76faa36233a9070cd4a5db9ee9b3f92323b602f"
channel = "stable"
[[nodes]]
name = "control-1"
role = "control"
address = "192.168.123.201"
hypervisor = "local"
failure_domain = "local-host"
mac_address = "52:54:00:67:01:01"
cpus = 2
memory_mib = 4096
disk_gib = 32
[[nodes]]
name = "worker-1"
role = "worker"
address = "192.168.123.202"
hypervisor = "local"
failure_domain = "local-host"
mac_address = "52:54:00:67:01:02"
cpus = 2
memory_mib = 4096
disk_gib = 40
[[nodes]]
name = "worker-2"
role = "worker"
address = "192.168.123.203"
hypervisor = "local"
failure_domain = "local-host"
mac_address = "52:54:00:67:01:03"
cpus = 2
memory_mib = 4096
disk_gib = 40
[[nodes]]
name = "state-1"
role = "state"
address = "192.168.123.204"
hypervisor = "local"
failure_domain = "local-host"
mac_address = "52:54:00:67:01:04"
cpus = 4
memory_mib = 4096
disk_gib = 80
+15
View File
@@ -0,0 +1,15 @@
#!/usr/bin/env python3
"""Executable entry point for the GovOPlaN Kubernetes VM lab."""
from __future__ import annotations
from pathlib import Path
import sys
sys.path.insert(0, str(Path(__file__).resolve().parent))
from govoplan_lab.cli import main
if __name__ == "__main__":
raise SystemExit(main())
+5
View File
@@ -0,0 +1,5 @@
"""Reproducible GovOPlaN Kubernetes acceptance lab."""
from .config import LabConfig, LabConfigError, LabNode, load_config
__all__ = ["LabConfig", "LabConfigError", "LabNode", "load_config"]
+153
View File
@@ -0,0 +1,153 @@
"""Command-line interface for the GovOPlaN Kubernetes VM lab."""
from __future__ import annotations
import argparse
from pathlib import Path
from typing import Sequence
from .config import LabConfigError, load_config
from .lifecycle import (
LabOperationError,
create,
deploy,
destroy,
doctor,
enroll_admin,
pause,
resume,
status,
update,
verify,
)
def build_parser() -> argparse.ArgumentParser:
parser = argparse.ArgumentParser(
prog="govoplan-lab",
description="Create and operate a libvirt-backed GovOPlaN Kubernetes test lab.",
)
parser.add_argument(
"--config",
type=Path,
default=Path("govoplan-lab.toml"),
help="Strict TOML lab inventory (default: ./govoplan-lab.toml).",
)
parser.add_argument("--verbose", action="store_true")
subparsers = parser.add_subparsers(dest="command", required=True)
doctor_parser = subparsers.add_parser("doctor", help="Validate inventory and prerequisites.")
doctor_parser.add_argument(
"--online",
action="store_true",
help="Also connect to every hypervisor and verify its toolchain.",
)
subparsers.add_parser("status", help="Show VM, Kubernetes node, and pod state.")
_mutation_parser(subparsers, "create", "Create or reuse all declared VMs.")
_mutation_parser(subparsers, "deploy", "Deploy shared state, K3s, and GovOPlaN.")
_mutation_parser(subparsers, "update", "Reconcile pinned K3s and GovOPlaN inputs serially.")
_mutation_parser(subparsers, "pause", "Gracefully stop the lab while preserving disks.")
_mutation_parser(subparsers, "resume", "Start a paused lab in dependency order.")
destroy_parser = _mutation_parser(
subparsers,
"destroy",
"Destroy lab-owned VMs and disks with an explicit name confirmation.",
)
destroy_parser.add_argument("--confirm", default="")
destroy_parser.add_argument(
"--purge-local-state",
action="store_true",
help="Also delete local secrets, manifests, and evidence after VM teardown.",
)
verify_parser = subparsers.add_parser(
"verify",
help=(
"Collect sanitized live-cluster evidence using an "
"Ops-read-authorized GOVOPLAN_OPS_API_KEY."
),
)
verify_parser.add_argument(
"--exercise-api-pod-loss",
action="store_true",
help="Delete one ready API pod during the bounded availability drill.",
)
enroll_parser = _mutation_parser(
subparsers,
"enroll-admin",
"Securely consume the first-administrator enrollment artifact.",
)
enroll_parser.add_argument("--email", required=True)
enroll_parser.add_argument("--display-name", default=None)
enroll_parser.add_argument("--tenant-slug", default="default")
enroll_parser.add_argument("--tenant-name", default="Default Tenant")
return parser
def _mutation_parser(
subparsers: argparse._SubParsersAction[argparse.ArgumentParser],
name: str,
help_text: str,
) -> argparse.ArgumentParser:
parser = subparsers.add_parser(name, help=help_text)
parser.add_argument(
"--apply",
action="store_true",
help="Perform mutations; without this flag the command is a dry run.",
)
return parser
def main(argv: Sequence[str] | None = None) -> int:
args = build_parser().parse_args(argv)
try:
config = load_config(args.config)
if args.command == "doctor":
return doctor(config, online=args.online, verbose=args.verbose)
if args.command == "status":
return status(config, verbose=args.verbose)
if args.command == "create":
create(config, apply=args.apply, verbose=args.verbose)
elif args.command == "deploy":
deploy(config, apply=args.apply, verbose=args.verbose)
elif args.command == "update":
update(config, apply=args.apply, verbose=args.verbose)
elif args.command == "pause":
pause(config, apply=args.apply, verbose=args.verbose)
elif args.command == "resume":
resume(config, apply=args.apply, verbose=args.verbose)
elif args.command == "destroy":
destroy(
config,
apply=args.apply,
confirmation=args.confirm,
purge_local_state=args.purge_local_state,
verbose=args.verbose,
)
elif args.command == "verify":
verify(
config,
exercise_api_pod_loss=args.exercise_api_pod_loss,
verbose=args.verbose,
)
elif args.command == "enroll-admin":
enroll_admin(
config,
email=args.email,
display_name=args.display_name,
tenant_slug=args.tenant_slug,
tenant_name=args.tenant_name,
apply=args.apply,
)
else:
raise RuntimeError(f"unsupported command: {args.command}")
return 0
except (LabConfigError, LabOperationError, OSError, ValueError) as exc:
print(f"error: {exc}", file=__import__("sys").stderr)
return 1
if __name__ == "__main__":
raise SystemExit(main())
+466
View File
@@ -0,0 +1,466 @@
"""Strict TOML model for the GovOPlaN Kubernetes VM lab."""
from __future__ import annotations
from dataclasses import dataclass
import ipaddress
from pathlib import Path
import re
import tomllib
from typing import Any, Mapping
from urllib.parse import urlsplit
SCHEMA_VERSION = 1
_NAME = re.compile(r"^[a-z][a-z0-9-]{1,47}$")
_HOSTNAME = re.compile(
r"^(?=.{1,253}$)(?:[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\.)*"
r"[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?$"
)
_SHA256 = re.compile(r"^[0-9a-f]{64}$")
_K3S_VERSION = re.compile(r"^v[0-9]+\.[0-9]+\.[0-9]+\+k3s[0-9]+$")
_SSH_TARGET = re.compile(r"^(?:[A-Za-z0-9_.-]+@)?[A-Za-z0-9_.:-]+$")
_BRIDGE = re.compile(r"^[A-Za-z0-9_.:-]{1,32}$")
_ROLE = {"control", "worker", "state"}
_MODE = {"rehearsal", "acceptance"}
class LabConfigError(ValueError):
"""Raised when the lab inventory cannot be used safely."""
@dataclass(frozen=True, slots=True)
class LabNode:
name: str
role: str
address: ipaddress.IPv4Address
hypervisor: str
failure_domain: str
mac_address: str
cpus: int
memory_mib: int
disk_gib: int
@property
def is_local(self) -> bool:
return self.hypervisor == "local"
@dataclass(frozen=True, slots=True)
class NetworkConfig:
prefix_length: int
gateway: ipaddress.IPv4Address
dns_servers: tuple[ipaddress.IPv4Address, ...]
bridge: str
@dataclass(frozen=True, slots=True)
class ImageConfig:
url: str
sha256: str
@dataclass(frozen=True, slots=True)
class K3sConfig:
version: str
binary_url: str
binary_sha256: str
install_script_url: str
install_script_sha256: str
@dataclass(frozen=True, slots=True)
class ReleaseConfig:
manifest_url: str
manifest_sha256: str
keyring_url: str
keyring_sha256: str
channel: str
@dataclass(frozen=True, slots=True)
class LabConfig:
source: Path
schema_version: int
name: str
mode: str
state_directory: Path
vm_image_directory: str
ssh_user: str
ssh_private_key: Path
ssh_public_key: Path
namespace: str
public_host: str
s3_host: str
ingress_class: str
module_set: str
api_replicas: int
web_replicas: int
worker_replicas: int
db_connection_limit: int
network: NetworkConfig
image: ImageConfig
k3s: K3sConfig
release: ReleaseConfig
nodes: tuple[LabNode, ...]
@property
def controls(self) -> tuple[LabNode, ...]:
return tuple(node for node in self.nodes if node.role == "control")
@property
def workers(self) -> tuple[LabNode, ...]:
return tuple(node for node in self.nodes if node.role == "worker")
@property
def state_node(self) -> LabNode:
return next(node for node in self.nodes if node.role == "state")
@property
def primary_control(self) -> LabNode:
return self.controls[0]
@property
def public_url(self) -> str:
return f"https://{self.public_host}"
@property
def s3_url(self) -> str:
return f"https://{self.s3_host}:9443"
@property
def evidence_capable(self) -> bool:
worker_domains = {node.failure_domain for node in self.workers}
worker_hypervisors = {node.hypervisor for node in self.workers}
return (
self.mode == "acceptance"
and len(worker_domains) == len(self.workers)
and len(worker_hypervisors) == len(self.workers)
and self.state_node.failure_domain not in worker_domains
and self.state_node.hypervisor not in worker_hypervisors
)
def load_config(path: Path) -> LabConfig:
source = path.expanduser().resolve()
try:
raw = tomllib.loads(source.read_text(encoding="utf-8"))
except FileNotFoundError as exc:
raise LabConfigError(f"lab configuration does not exist: {source}") from exc
except tomllib.TOMLDecodeError as exc:
raise LabConfigError(f"lab configuration is not valid TOML: {exc}") from exc
root = _mapping(raw, "lab configuration")
_only_keys(
root,
{
"schema_version",
"name",
"mode",
"state_directory",
"vm_image_directory",
"ssh_user",
"ssh_private_key",
"ssh_public_key",
"namespace",
"public_host",
"s3_host",
"ingress_class",
"module_set",
"api_replicas",
"web_replicas",
"worker_replicas",
"db_connection_limit",
"network",
"image",
"k3s",
"release",
"nodes",
},
"lab configuration",
)
schema_version = _integer(root, "schema_version")
if schema_version != SCHEMA_VERSION:
raise LabConfigError(
f"schema_version must be {SCHEMA_VERSION}; found {schema_version}"
)
name = _pattern(root, "name", _NAME)
mode = _choice(root, "mode", _MODE)
base = source.parent
state_directory = _path(root, "state_directory", base)
vm_image_directory = _absolute_posix_path(root, "vm_image_directory")
ssh_user = _pattern(root, "ssh_user", re.compile(r"^[a-z_][a-z0-9_-]{0,31}$"))
ssh_private_key = _path(root, "ssh_private_key", base)
ssh_public_key = _path(root, "ssh_public_key", base)
namespace = _pattern(root, "namespace", _NAME)
public_host = _pattern(root, "public_host", _HOSTNAME)
s3_host = _pattern(root, "s3_host", _HOSTNAME)
if public_host == s3_host:
raise LabConfigError("public_host and s3_host must be different")
ingress_class = _pattern(root, "ingress_class", _NAME)
module_set = _choice(root, "module_set", {"core", "base", "full"})
api_replicas = _bounded_integer(root, "api_replicas", 2, 32)
web_replicas = _bounded_integer(root, "web_replicas", 2, 32)
worker_replicas = _bounded_integer(root, "worker_replicas", 2, 64)
db_connection_limit = _bounded_integer(root, "db_connection_limit", 50, 10000)
network = _parse_network(_mapping(root.get("network"), "network"))
image = _parse_image(_mapping(root.get("image"), "image"))
k3s = _parse_k3s(_mapping(root.get("k3s"), "k3s"))
release = _parse_release(_mapping(root.get("release"), "release"))
raw_nodes = root.get("nodes")
if not isinstance(raw_nodes, list) or not raw_nodes:
raise LabConfigError("nodes must be a non-empty array of tables")
nodes = tuple(_parse_node(item, index=index) for index, item in enumerate(raw_nodes))
config = LabConfig(
source=source,
schema_version=schema_version,
name=name,
mode=mode,
state_directory=state_directory,
vm_image_directory=vm_image_directory.rstrip("/"),
ssh_user=ssh_user,
ssh_private_key=ssh_private_key,
ssh_public_key=ssh_public_key,
namespace=namespace,
public_host=public_host,
s3_host=s3_host,
ingress_class=ingress_class,
module_set=module_set,
api_replicas=api_replicas,
web_replicas=web_replicas,
worker_replicas=worker_replicas,
db_connection_limit=db_connection_limit,
network=network,
image=image,
k3s=k3s,
release=release,
nodes=nodes,
)
_validate_topology(config)
return config
def _parse_network(raw: Mapping[str, Any]) -> NetworkConfig:
_only_keys(raw, {"prefix_length", "gateway", "dns_servers", "bridge"}, "network")
prefix_length = _bounded_integer(raw, "prefix_length", 8, 30)
gateway = _ipv4(raw, "gateway")
dns_raw = raw.get("dns_servers")
if not isinstance(dns_raw, list) or not dns_raw or len(dns_raw) > 4:
raise LabConfigError("network.dns_servers must contain 1-4 IPv4 addresses")
dns_servers = tuple(_ipv4_value(value, "network.dns_servers") for value in dns_raw)
bridge = _pattern(raw, "bridge", _BRIDGE)
return NetworkConfig(prefix_length, gateway, dns_servers, bridge)
def _parse_image(raw: Mapping[str, Any]) -> ImageConfig:
_only_keys(raw, {"url", "sha256"}, "image")
return ImageConfig(
url=_https_url(raw, "url"),
sha256=_pattern(raw, "sha256", _SHA256),
)
def _parse_k3s(raw: Mapping[str, Any]) -> K3sConfig:
_only_keys(
raw,
{
"version",
"binary_url",
"binary_sha256",
"install_script_url",
"install_script_sha256",
},
"k3s",
)
return K3sConfig(
version=_pattern(raw, "version", _K3S_VERSION),
binary_url=_https_url(raw, "binary_url"),
binary_sha256=_pattern(raw, "binary_sha256", _SHA256),
install_script_url=_https_url(raw, "install_script_url"),
install_script_sha256=_pattern(raw, "install_script_sha256", _SHA256),
)
def _parse_release(raw: Mapping[str, Any]) -> ReleaseConfig:
_only_keys(
raw,
{"manifest_url", "manifest_sha256", "keyring_url", "keyring_sha256", "channel"},
"release",
)
return ReleaseConfig(
manifest_url=_https_url(raw, "manifest_url"),
manifest_sha256=_pattern(raw, "manifest_sha256", _SHA256),
keyring_url=_https_url(raw, "keyring_url"),
keyring_sha256=_pattern(raw, "keyring_sha256", _SHA256),
channel=_pattern(raw, "channel", _NAME),
)
def _parse_node(value: object, *, index: int) -> LabNode:
raw = _mapping(value, f"nodes[{index}]")
_only_keys(
raw,
{
"name",
"role",
"address",
"hypervisor",
"failure_domain",
"mac_address",
"cpus",
"memory_mib",
"disk_gib",
},
f"nodes[{index}]",
)
hypervisor = _string(raw, "hypervisor")
if hypervisor != "local" and not _SSH_TARGET.fullmatch(hypervisor):
raise LabConfigError(
f"nodes[{index}].hypervisor must be 'local' or a simple SSH target"
)
mac_address = _string(raw, "mac_address").lower()
try:
octets = mac_address.split(":")
valid_mac = len(octets) == 6 and all(
len(octet) == 2 and 0 <= int(octet, 16) <= 255 for octet in octets
)
except ValueError:
valid_mac = False
if not valid_mac:
raise LabConfigError(f"nodes[{index}].mac_address is not a canonical MAC address")
return LabNode(
name=_pattern(raw, "name", _NAME),
role=_choice(raw, "role", _ROLE),
address=_ipv4(raw, "address"),
hypervisor=hypervisor,
failure_domain=_pattern(raw, "failure_domain", _NAME),
mac_address=mac_address,
cpus=_bounded_integer(raw, "cpus", 1, 64),
memory_mib=_bounded_integer(raw, "memory_mib", 2048, 262144),
disk_gib=_bounded_integer(raw, "disk_gib", 16, 4096),
)
def _validate_topology(config: LabConfig) -> None:
names = [node.name for node in config.nodes]
addresses = [node.address for node in config.nodes]
mac_addresses = [node.mac_address for node in config.nodes]
for label, values in (
("node names", names),
("node addresses", addresses),
("node MAC addresses", mac_addresses),
):
if len(values) != len(set(values)):
raise LabConfigError(f"{label} must be unique")
too_long = [
node.name
for node in config.nodes
if len(f"{config.name}-{node.name}") > 63
]
if too_long:
raise LabConfigError(
"lab name plus node name must fit the 63-character libvirt domain limit: "
+ ", ".join(too_long)
)
if len(config.controls) not in {1, 3}:
raise LabConfigError("the lab requires exactly one or three control-plane nodes")
if len(config.workers) < 2:
raise LabConfigError("the lab requires at least two worker nodes")
if sum(node.role == "state" for node in config.nodes) != 1:
raise LabConfigError("the lab requires exactly one external shared-state node")
network = ipaddress.ip_network(
f"{config.network.gateway}/{config.network.prefix_length}", strict=False
)
if any(node.address not in network for node in config.nodes):
raise LabConfigError("every node address must be in the configured IPv4 network")
if config.network.gateway in addresses:
raise LabConfigError("the network gateway cannot also be a node address")
if config.mode == "acceptance" and not config.evidence_capable:
raise LabConfigError(
"acceptance mode requires each worker and the shared-state node to use "
"distinct hypervisors and failure_domain values"
)
def _mapping(value: object, label: str) -> Mapping[str, Any]:
if not isinstance(value, dict):
raise LabConfigError(f"{label} must be a table")
return value
def _only_keys(raw: Mapping[str, Any], allowed: set[str], label: str) -> None:
unexpected = sorted(set(raw) - allowed)
if unexpected:
raise LabConfigError(f"{label} contains unsupported keys: {', '.join(unexpected)}")
def _string(raw: Mapping[str, Any], key: str) -> str:
value = raw.get(key)
if not isinstance(value, str) or not value.strip():
raise LabConfigError(f"{key} must be a non-empty string")
return value.strip()
def _integer(raw: Mapping[str, Any], key: str) -> int:
value = raw.get(key)
if not isinstance(value, int) or isinstance(value, bool):
raise LabConfigError(f"{key} must be an integer")
return value
def _bounded_integer(raw: Mapping[str, Any], key: str, minimum: int, maximum: int) -> int:
value = _integer(raw, key)
if not minimum <= value <= maximum:
raise LabConfigError(f"{key} must be between {minimum} and {maximum}")
return value
def _pattern(raw: Mapping[str, Any], key: str, pattern: re.Pattern[str]) -> str:
value = _string(raw, key)
if pattern.fullmatch(value) is None:
raise LabConfigError(f"{key} has an unsupported format")
return value
def _choice(raw: Mapping[str, Any], key: str, choices: set[str]) -> str:
value = _string(raw, key)
if value not in choices:
raise LabConfigError(f"{key} must be one of: {', '.join(sorted(choices))}")
return value
def _ipv4(raw: Mapping[str, Any], key: str) -> ipaddress.IPv4Address:
return _ipv4_value(_string(raw, key), key)
def _ipv4_value(value: object, label: str) -> ipaddress.IPv4Address:
if not isinstance(value, str):
raise LabConfigError(f"{label} must contain strings")
try:
parsed = ipaddress.ip_address(value)
except ValueError as exc:
raise LabConfigError(f"{label} contains an invalid IP address") from exc
if not isinstance(parsed, ipaddress.IPv4Address):
raise LabConfigError(f"{label} supports IPv4 only in schema version 1")
return parsed
def _path(raw: Mapping[str, Any], key: str, base: Path) -> Path:
value = Path(_string(raw, key)).expanduser()
return (value if value.is_absolute() else base / value).resolve()
def _absolute_posix_path(raw: Mapping[str, Any], key: str) -> str:
value = _string(raw, key)
if not value.startswith("/") or ".." in Path(value).parts:
raise LabConfigError(f"{key} must be an absolute path without '..'")
return value
def _https_url(raw: Mapping[str, Any], key: str) -> str:
value = _string(raw, key)
parsed = urlsplit(value)
if parsed.scheme != "https" or not parsed.hostname or parsed.username or parsed.password:
raise LabConfigError(f"{key} must be a credential-free HTTPS URL")
if parsed.fragment:
raise LabConfigError(f"{key} must not contain a fragment")
return value
File diff suppressed because it is too large Load Diff
+311
View File
@@ -0,0 +1,311 @@
"""Deterministic configuration rendering for the Kubernetes VM lab."""
from __future__ import annotations
import base64
import json
from pathlib import Path
from typing import Mapping
from .config import LabConfig, LabNode
def render_cloud_init(config: LabConfig, node: LabNode, public_key: str) -> str:
packages = ["ca-certificates", "curl", "qemu-guest-agent"]
if node.role == "state":
packages.extend(["docker.io", "docker-compose-v2", "openssl"])
package_lines = "\n".join(f" - {item}" for item in packages)
return f"""#cloud-config
hostname: {node.name}
manage_etc_hosts: true
package_update: true
package_upgrade: false
packages:
{package_lines}
users:
- default
- name: {config.ssh_user}
groups: [adm, sudo]
shell: /bin/bash
sudo: ALL=(ALL) NOPASSWD:ALL
lock_passwd: true
ssh_authorized_keys:
- {json.dumps(public_key.strip())}
ssh_pwauth: false
disable_root: true
runcmd:
- [systemctl, enable, --now, qemu-guest-agent]
- [sh, -c, "test ! -e /usr/bin/docker || systemctl enable --now docker"]
final_message: "GovOPlaN lab node is ready"
"""
def render_network_config(config: LabConfig, node: LabNode) -> str:
dns = ", ".join(str(item) for item in config.network.dns_servers)
return f"""version: 2
ethernets:
primary:
match:
macaddress: {node.mac_address}
set-name: eth0
addresses:
- {node.address}/{config.network.prefix_length}
routes:
- to: default
via: {config.network.gateway}
nameservers:
addresses: [{dns}]
"""
def render_meta_data(config: LabConfig, node: LabNode) -> str:
return f"instance-id: {config.name}-{node.name}\nlocal-hostname: {node.name}\n"
def render_k3s_config(
config: LabConfig,
node: LabNode,
*,
cluster_token: str,
) -> str:
lines = [
f'node-name: "{node.name}"',
f'node-ip: "{node.address}"',
f'token: "{cluster_token}"',
]
if node.role == "control":
if node == config.primary_control:
lines.append("cluster-init: true")
else:
lines.append(f'server: "https://{config.primary_control.address}:6443"')
lines.extend(
[
'write-kubeconfig-mode: "0600"',
"secrets-encryption: true",
"tls-san:",
f' - "{config.primary_control.address}"',
"node-taint:",
' - "node-role.kubernetes.io/control-plane=true:NoSchedule"',
]
)
elif node.role == "worker":
lines.extend(
[
f'server: "https://{config.primary_control.address}:6443"',
"node-label:",
' - "govoplan.add-ideas.de/runtime=true"',
f' - "topology.govoplan.add-ideas.de/failure-domain={node.failure_domain}"',
]
)
else:
raise ValueError("state nodes do not receive K3s configuration")
return "\n".join(lines) + "\n"
def render_registry_config(username: str, password: str) -> str:
if not username and not password:
return ""
if not username or not password:
raise ValueError("registry username and password must be supplied together")
return (
'mirrors:\n "git.add-ideas.de":\n'
' endpoint:\n - "https://git.add-ideas.de"\n'
'configs:\n "git.add-ideas.de":\n auth:\n'
f" username: {json.dumps(username)}\n"
f" password: {json.dumps(password)}\n"
)
def render_garage_config() -> str:
return """metadata_dir = "/var/lib/garage/meta"
data_dir = "/var/lib/garage/data"
db_engine = "sqlite"
replication_factor = 1
rpc_bind_addr = "[::]:3901"
rpc_public_addr = "127.0.0.1:3901"
[s3_api]
s3_region = "garage"
api_bind_addr = "[::]:3900"
root_domain = ".s3.garage.localhost"
[admin]
api_bind_addr = "[::]:3903"
"""
def render_caddyfile() -> str:
return """:9443 {
tls /etc/caddy/tls/server.crt /etc/caddy/tls/server.key
reverse_proxy garage:3900
}
"""
def render_state_compose(images: Mapping[str, str]) -> str:
required = {"postgres", "redis", "garage", "managed_ingress", "test_mail"}
missing = sorted(required - set(images))
if missing:
raise ValueError("release manifest is missing state images: " + ", ".join(missing))
value = {
"name": "govoplan-lab-state",
"services": {
"postgres": {
"image": images["postgres"],
"restart": "unless-stopped",
"environment": {
"POSTGRES_DB": "${POSTGRES_DB}",
"POSTGRES_USER": "${POSTGRES_USER}",
"POSTGRES_PASSWORD": "${POSTGRES_PASSWORD}",
},
"ports": ["5432:5432"],
"healthcheck": {
"test": [
"CMD-SHELL",
'pg_isready -U "$${POSTGRES_USER}" -d "$${POSTGRES_DB}"',
],
"interval": "5s",
"timeout": "3s",
"retries": 30,
},
"volumes": ["postgres-data:/var/lib/postgresql/data"],
"networks": ["internal"],
},
"redis": {
"image": images["redis"],
"restart": "unless-stopped",
"command": [
"sh",
"-ec",
'exec redis-server --appendonly yes --requirepass "$$REDIS_PASSWORD"',
],
"environment": {"REDIS_PASSWORD": "${REDIS_PASSWORD}"},
"ports": ["6379:6379"],
"healthcheck": {
"test": [
"CMD-SHELL",
'redis-cli -a "$${REDIS_PASSWORD}" --no-auth-warning ping',
],
"interval": "5s",
"timeout": "3s",
"retries": 30,
},
"volumes": ["redis-data:/data"],
"networks": ["internal"],
},
"garage": {
"image": images["garage"],
"restart": "unless-stopped",
"command": ["/garage", "server", "--single-node", "--default-bucket"],
"environment": {
"GARAGE_DEFAULT_ACCESS_KEY": "${GARAGE_DEFAULT_ACCESS_KEY}",
"GARAGE_DEFAULT_SECRET_KEY": "${GARAGE_DEFAULT_SECRET_KEY}",
"GARAGE_DEFAULT_BUCKET": "${GARAGE_DEFAULT_BUCKET}",
"GARAGE_RPC_SECRET": "${GARAGE_RPC_SECRET}",
"GARAGE_ADMIN_TOKEN": "${GARAGE_ADMIN_TOKEN}",
"GARAGE_METRICS_TOKEN": "${GARAGE_METRICS_TOKEN}",
},
"healthcheck": {
"test": ["CMD", "/garage", "status"],
"interval": "10s",
"timeout": "5s",
"retries": 30,
"start_period": "15s",
},
"security_opt": ["no-new-privileges:true"],
"volumes": [
"./garage.toml:/etc/garage.toml:ro",
"garage-meta:/var/lib/garage/meta",
"garage-data:/var/lib/garage/data",
],
"networks": ["internal"],
},
"s3-tls": {
"image": images["managed_ingress"],
"restart": "unless-stopped",
"depends_on": {"garage": {"condition": "service_healthy"}},
"ports": ["9443:9443"],
"volumes": [
"./Caddyfile:/etc/caddy/Caddyfile:ro",
"./tls:/etc/caddy/tls:ro",
],
"security_opt": ["no-new-privileges:true"],
"networks": ["internal"],
},
"test-mail": {
"image": images["test_mail"],
"restart": "unless-stopped",
"environment": {
"GREENMAIL_OPTS": (
"-Dgreenmail.setup.test.smtp -Dgreenmail.setup.test.imap "
"-Dgreenmail.hostname=0.0.0.0"
)
},
"ports": ["3025:3025", "3143:3143"],
"networks": ["internal"],
},
},
"volumes": {
"postgres-data": {},
"redis-data": {},
"garage-meta": {},
"garage-data": {},
},
"networks": {"internal": {"driver": "bridge"}},
}
return json.dumps(value, indent=2, sort_keys=True) + "\n"
def render_state_environment(values: Mapping[str, str]) -> str:
return "".join(f"{key}={_env_quote(value)}\n" for key, value in sorted(values.items()))
def render_secret_manifest(
*, namespace: str, name: str, values: Mapping[str, bytes | str]
) -> bytes:
encoded = {
key: base64.b64encode(value.encode() if isinstance(value, str) else value).decode()
for key, value in sorted(values.items())
}
payload = {
"apiVersion": "v1",
"kind": "Secret",
"metadata": {"name": name, "namespace": namespace},
"type": "Opaque",
"data": encoded,
}
return (json.dumps(payload, indent=2, sort_keys=True) + "\n").encode()
def render_tls_secret_manifest(
*, namespace: str, name: str, certificate: bytes, private_key: bytes
) -> bytes:
payload = json.loads(
render_secret_manifest(
namespace=namespace,
name=name,
values={"tls.crt": certificate, "tls.key": private_key},
)
)
payload["type"] = "kubernetes.io/tls"
return (json.dumps(payload, indent=2, sort_keys=True) + "\n").encode()
def render_hosts(config: LabConfig) -> str:
return (
f"{config.primary_control.address} {config.public_host}\n"
f"{config.state_node.address} {config.s3_host}\n"
)
def _env_quote(value: str) -> str:
return json.dumps(value, ensure_ascii=True)
def write_private(path: Path, value: str | bytes) -> None:
path.parent.mkdir(parents=True, exist_ok=True, mode=0o700)
data = value.encode() if isinstance(value, str) else value
path.write_bytes(data)
path.chmod(0o600)
@@ -0,0 +1,11 @@
<network>
<name>govoplan-lab</name>
<forward mode="nat"/>
<bridge name="virbr-gplab" stp="on" delay="0"/>
<domain name="lab.test" localOnly="yes"/>
<ip address="192.168.123.1" netmask="255.255.255.0">
<dhcp>
<range start="192.168.123.2" end="192.168.123.99"/>
</dhcp>
</ip>
</network>