DevOpsTools/proxmox-api-simulator
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Prepare 0.1.0 for lab release: durable handlers, HTTP Compose, CI, and pulumi-tests.

Browse Source 
- Harden DB-backed handlers and seed profiles; align client wire shapes for
  cluster resources, QEMU config, and node SSL fields
- Serve plain HTTP on Compose :8006; keep TLS optional (--profile tls) and
  terminate HTTPS at Kubernetes Ingress
- Add pulumi-tests (full contract surface majors 6–9 + BPG lifecycle) and
  make pulumi-tests
- Ship bilingual docs, CHANGELOG, SECURITY, CONTRIBUTING, and GitHub Actions
  (make ci + Compose/Helm validation)
This commit is contained in:
Sergey Antropoff
2026-07-18 04:18:05 +03:00
parent 777926487b
commit 48df10b17e
172 changed files with 7528 additions and 1208 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/handlers/cluster_extra.py
+60 -96
View File
@@ -2,7 +2,6 @@
from __future__ import annotations
import copy
import secrets
import time
from typing import Any
@@ -22,56 +21,20 @@ from app.handlers.common import (
from app.tasks.repository import TaskRepository
from app.tasks.upid import Upid
DEFAULT_CEPH_FLAGS: dict[str, int] = {
"nobackfill": 0,
"nodeep-scrub": 0,
"nodown": 0,
"noin": 0,
"noout": 0,
"norebalance": 0,
"norecover": 0,
"noscrub": 0,
"notieragent": 0,
"pause": 0,
}
DEFAULT_CPU_FLAGS: list[dict[str, Any]] = [
{"name": "aes", "introduces": "Westmere"},
{"name": "avx", "introduces": "SandyBridge"},
{"name": "avx2", "introduces": "Haswell"},
]
def _jobs(metadata: dict[str, Any]) -> dict[str, Any]:
jobs = metadata.setdefault("jobs", {})
if not isinstance(jobs, dict):
jobs = {}
metadata["jobs"] = jobs
sync = jobs.setdefault("realm_sync", {})
if not isinstance(sync, dict):
sync = {}
jobs["realm_sync"] = sync
return jobs
jobs = metadata.get("jobs")
return jobs if isinstance(jobs, dict) else {}
def _metrics(metadata: dict[str, Any]) -> dict[str, Any]:
metrics = metadata.setdefault("metrics", {})
if not isinstance(metrics, dict):
metrics = {}
metadata["metrics"] = metrics
servers = metrics.setdefault("servers", {})
if not isinstance(servers, dict):
servers = {}
metrics["servers"] = servers
return metrics
metrics = metadata.get("metrics")
return metrics if isinstance(metrics, dict) else {}
def _cpu_models(metadata: dict[str, Any]) -> dict[str, Any]:
models = metadata.get("qemu_cpu_models")
if not isinstance(models, dict):
models = {}
metadata["qemu_cpu_models"] = models
return models
return models if isinstance(models, dict) else {}
def _ha_rules_store(metadata: dict[str, Any]) -> list[dict[str, Any]]:
@@ -84,12 +47,7 @@ def _ha_rules_store(metadata: dict[str, Any]) -> list[dict[str, Any]]:
]
if isinstance(rules, list):
return [dict(item) for item in rules if isinstance(item, dict)]
defaults = [
{"rule": "node-fencing", "type": "node", "action": "restart"},
{"rule": "service-ha", "type": "resource", "action": "failover"},
]
metadata["ha_rules"] = defaults
return list(defaults)
return []
def _save_ha_rules(metadata: dict[str, Any], rules: list[dict[str, Any]]) -> None:
@@ -105,18 +63,7 @@ def _replication_jobs(metadata: dict[str, Any]) -> list[dict[str, Any]]:
def _ceph(metadata: dict[str, Any]) -> dict[str, Any]:
ceph = metadata.get("ceph")
if not isinstance(ceph, dict):
ceph = {}
flags = ceph.get("flags")
if not isinstance(flags, dict):
flags = copy.deepcopy(DEFAULT_CEPH_FLAGS)
else:
merged = copy.deepcopy(DEFAULT_CEPH_FLAGS)
merged.update({str(key): int(value) for key, value in flags.items()})
flags = merged
ceph["flags"] = flags
metadata["ceph"] = ceph
return ceph
return ceph if isinstance(ceph, dict) else {}
async def _cluster_task(request: Request, *, task_type: str, worker: str) -> str:
@@ -174,8 +121,8 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
payload = values(inputs)
job_id = str(payload["id"])
metadata = await cluster_metadata(request)
jobs = _jobs(metadata)
sync = jobs.setdefault("realm_sync", {})
jobs = dict(_jobs(metadata))
sync = dict(jobs.get("realm_sync") or {})
if job_id in sync:
raise ApiError(409, "realm-sync job already exists")
entry = {
@@ -185,6 +132,7 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
entry.setdefault("enabled", 1)
entry.setdefault("realm", str(payload.get("realm") or "pam"))
sync[job_id] = entry
jobs["realm_sync"] = sync
metadata["jobs"] = jobs
await save_cluster_metadata(request, metadata)
return {"id": job_id, **entry}
@@ -193,8 +141,8 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
payload = values(inputs)
job_id = str(payload["id"])
metadata = await cluster_metadata(request)
jobs = _jobs(metadata)
sync = jobs.setdefault("realm_sync", {})
jobs = dict(_jobs(metadata))
sync = dict(jobs.get("realm_sync") or {})
if job_id not in sync:
raise ApiError(404, "realm-sync job does not exist")
updated = {
@@ -206,6 +154,7 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
},
}
sync[job_id] = updated
jobs["realm_sync"] = sync
metadata["jobs"] = jobs
await save_cluster_metadata(request, metadata)
return {"id": job_id, **updated}
@@ -213,23 +162,26 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
async def realm_sync_delete(request: Request, inputs: dict[str, Any]) -> None:
job_id = str(values(inputs)["id"])
metadata = await cluster_metadata(request)
jobs = _jobs(metadata)
sync = jobs.setdefault("realm_sync", {})
jobs = dict(_jobs(metadata))
sync = dict(jobs.get("realm_sync") or {})
if job_id not in sync:
raise ApiError(404, "realm-sync job does not exist")
del sync[job_id]
jobs["realm_sync"] = sync
metadata["jobs"] = jobs
await save_cluster_metadata(request, metadata)
async def schedule_analyze(request: Request, inputs: dict[str, Any]) -> list[dict[str, Any]]:
schedule = str(values(inputs).get("schedule") or "*/15")
metadata = await cluster_metadata(request)
jobs = _jobs(metadata)
jobs = dict(_jobs(metadata))
results = jobs.get("schedule_analyze_results")
if not isinstance(results, list):
results = []
jobs["last_schedule_analyze"] = {"schedule": schedule, "at": int(time.time())}
metadata["jobs"] = jobs
await save_cluster_metadata(request, metadata)
now = int(time.time())
return [{"timestamp": now + offset * 900, "utc": True} for offset in range(4)]
return [dict(item) for item in results if isinstance(item, dict)]
async def metrics_index(_request: Request, _inputs: dict[str, Any]) -> list[dict[str, str]]:
return subdirs("export", "server")
@@ -238,8 +190,7 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
metadata = await cluster_metadata(request)
metrics = _metrics(metadata)
return {
"data": metrics.get("export_data")
or '# HELP pve_up Node is up\npve_up{node="pve01"} 1\n',
"data": str(metrics.get("export_data") or ""),
"timestamp": int(time.time()),
}
@@ -248,9 +199,12 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
) -> list[dict[str, Any]]:
metadata = await cluster_metadata(request)
metrics = _metrics(metadata)
servers = metrics.get("servers")
if not isinstance(servers, dict):
return []
return [
{"id": server_id, **dict(payload)}
for server_id, payload in sorted(metrics.get("servers", {}).items())
for server_id, payload in sorted(servers.items())
if isinstance(payload, dict)
]
@@ -267,8 +221,8 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
payload = values(inputs)
server_id = str(payload["id"])
metadata = await cluster_metadata(request)
metrics = _metrics(metadata)
servers = metrics.setdefault("servers", {})
metrics = dict(_metrics(metadata))
servers = dict(metrics.get("servers") or {})
if server_id in servers:
raise ApiError(409, "metrics server already exists")
entry = {
@@ -279,6 +233,7 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
entry.setdefault("port", 8086)
entry.setdefault("enable", 1)
servers[server_id] = entry
metrics["servers"] = servers
metadata["metrics"] = metrics
await save_cluster_metadata(request, metadata)
return {"id": server_id, **entry}
@@ -287,8 +242,8 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
payload = values(inputs)
server_id = str(payload["id"])
metadata = await cluster_metadata(request)
metrics = _metrics(metadata)
servers = metrics.setdefault("servers", {})
metrics = dict(_metrics(metadata))
servers = dict(metrics.get("servers") or {})
if server_id not in servers:
raise ApiError(404, "metrics server does not exist")
updated = {
@@ -300,6 +255,7 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
},
}
servers[server_id] = updated
metrics["servers"] = servers
metadata["metrics"] = metrics
await save_cluster_metadata(request, metadata)
return {"id": server_id, **updated}
@@ -307,19 +263,24 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
async def metrics_server_delete(request: Request, inputs: dict[str, Any]) -> None:
server_id = str(values(inputs)["id"])
metadata = await cluster_metadata(request)
metrics = _metrics(metadata)
servers = metrics.setdefault("servers", {})
metrics = dict(_metrics(metadata))
servers = dict(metrics.get("servers") or {})
if server_id not in servers:
raise ApiError(404, "metrics server does not exist")
del servers[server_id]
metrics["servers"] = servers
metadata["metrics"] = metrics
await save_cluster_metadata(request, metadata)
async def qemu_index(_request: Request, _inputs: dict[str, Any]) -> list[dict[str, str]]:
return subdirs("cpu-flags", "custom-cpu-models")
async def qemu_cpu_flags(_request: Request, _inputs: dict[str, Any]) -> list[dict[str, Any]]:
return list(DEFAULT_CPU_FLAGS)
async def qemu_cpu_flags(request: Request, _inputs: dict[str, Any]) -> list[dict[str, Any]]:
metadata = await cluster_metadata(request)
flags = metadata.get("qemu_cpu_flags")
if isinstance(flags, list):
return [dict(item) for item in flags if isinstance(item, dict)]
return []
async def cpu_models_list(request: Request, _inputs: dict[str, Any]) -> list[dict[str, Any]]:
metadata = await cluster_metadata(request)
@@ -440,14 +401,16 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
async def ceph_flags_get(request: Request, _inputs: dict[str, Any]) -> dict[str, int]:
metadata = await cluster_metadata(request)
ceph = _ceph(metadata)
await save_cluster_metadata(request, metadata)
return {str(key): int(value) for key, value in ceph["flags"].items()}
flags = ceph.get("flags")
if not isinstance(flags, dict):
return {}
return {str(key): int(value) for key, value in flags.items()}
async def ceph_flags_put(request: Request, inputs: dict[str, Any]) -> dict[str, int]:
payload = values(inputs)
metadata = await cluster_metadata(request)
ceph = _ceph(metadata)
flags = dict(ceph["flags"])
ceph = dict(_ceph(metadata))
flags = dict(ceph.get("flags") or {})
for key, value in payload.items():
if key in {"delete", "digest"}:
continue
@@ -461,8 +424,8 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
flag = str(values(inputs)["flag"])
metadata = await cluster_metadata(request)
ceph = _ceph(metadata)
flags = ceph["flags"]
if flag not in flags:
flags = ceph.get("flags")
if not isinstance(flags, dict) or flag not in flags:
raise ApiError(404, "ceph flag does not exist")
return {flag: int(flags[flag])}
@@ -470,8 +433,8 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
payload = values(inputs)
flag = str(payload["flag"])
metadata = await cluster_metadata(request)
ceph = _ceph(metadata)
flags = dict(ceph["flags"])
ceph = dict(_ceph(metadata))
flags = dict(ceph.get("flags") or {})
if "value" in payload:
flags[flag] = int(payload["value"])
elif flag in payload:
@@ -486,14 +449,15 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
async def ceph_metadata(request: Request, _inputs: dict[str, Any]) -> dict[str, Any]:
metadata = await cluster_metadata(request)
ceph = _ceph(metadata)
await save_cluster_metadata(request, metadata)
config_raw = ceph.get("config")
config: dict[str, Any] = dict(config_raw) if isinstance(config_raw, dict) else {}
version = ceph.get("version")
flags = ceph.get("flags")
return {
"version": ceph.get("version") or {"str": "18.2.2", "parts": [18, 2, 2]},
"fsid": ceph.get("config", {}).get("fsid")
if isinstance(ceph.get("config"), dict)
else "pve-simulator-fsid",
"initialized": int(bool(ceph.get("initialized", True))),
"flags": ceph.get("flags", {}),
"version": dict(version) if isinstance(version, dict) else {},
"fsid": str(config.get("fsid") or ""),
"initialized": int(bool(ceph.get("initialized"))),
"flags": dict(flags) if isinstance(flags, dict) else {},
}
async def ha_rule_create(request: Request, inputs: dict[str, Any]) -> None:
@@ -588,7 +552,7 @@ def register_cluster_extra_handlers(registry: HandlerRegistry) -> None:
log = job.get("log")
if isinstance(log, list):
return [dict(item) for item in log if isinstance(item, dict)]
return [{"t": int(time.time()), "n": 0, "msg": f"replication idle for {job.get('id')}"}]
return []
async def node_replication_status(request: Request, inputs: dict[str, Any]) -> dict[str, Any]:
job = await node_replication_get(request, inputs)