A review of the backup path found several ways a backup was not one, and a restore could destroy data it did not contain. Data loss: - Restore with overwrite rmtree'd the whole stack directory, including folders the backup had left out (over 2 GiB, deselected, an NFS share mounted inside). It now clears only what the backup holds, keeps the excluded paths (recorded in the manifest) and never crosses into another filesystem. - The stack was only stopped when named volumes were selected, so data folders inside the stack directory were archived while in use. Any selected data now stops it. - Retention matched by prefix: pruning "app" deleted "app-db"'s backups. It now matches backup-<id>-(full|config)-<date>.tar.gz exactly. - A restore under a new name wrote into the original's volumes when they had an explicit name, and into its folders outside the stack directory. Both are skipped now; a volume owned by another stack or mounted by a running container is never overwritten. - A truncated archive was only noticed after the old stack was gone. The whole archive is verified first, and the stack directory is unpacked next to the old one and swapped in. Remote storage: - Bind folders on a mounted share (fstab/autofs), bind-type volumes pointing at one and shares mounted inside the stack folder are detected by filesystem type and not selected; a folder du cannot measure in 20s is no longer selected by default either. Reliability: - After a backup exactly the containers that ran are started again with compose start (falling back to the Docker API), instead of up -d, which started services stopped on purpose and applied undeployed edits. A stack that cannot be stopped is not backed up. - Backup and restore hold the stack lock; a scheduled backup that finds the stack busy retries after 10 minutes. Restore stops the stack while replacing its data and brings it back up. - external: true volumes are backed up; anonymous ones are listed, not selected. Restored volumes get their original driver and options. - A backup with uncaptured data or a stack that did not restart is reported as incomplete: in the dialog, and for schedules as a warning with a notification and without rotating older backups out. - Archive writing and restore run off the event loop; downloads delete their temp file; helper containers use the long-timeout client; S3 listing is paginated; a file changing mid-read no longer corrupts the archive. Manifest format 3 records compose_binds/excluded and volume driver/options/external; format 1/2 backups still restore. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
496 lines
17 KiB
Python
496 lines
17 KiB
Python
"""File-based stack storage and Docker Compose lifecycle.
|
||
|
||
The compose YAML on disk is always the source of truth. The database only
|
||
stores metadata (name, description, timestamps).
|
||
"""
|
||
from __future__ import annotations
|
||
|
||
import asyncio
|
||
import os
|
||
import re
|
||
import shutil
|
||
import tempfile
|
||
from dataclasses import dataclass, field
|
||
from datetime import datetime, timezone
|
||
from typing import Optional
|
||
|
||
from config import settings
|
||
from docker_client import DockerError, get_client, safe_call
|
||
from services import registry_service
|
||
|
||
COMPOSE_FILENAMES = ("compose.yaml", "compose.yml", "docker-compose.yml", "docker-compose.yaml")
|
||
DEFAULT_COMPOSE_NAME = "compose.yaml"
|
||
COMPOSE_LABEL = "com.docker.compose.project"
|
||
SERVICE_LABEL = "com.docker.compose.service"
|
||
|
||
|
||
class StackFileError(Exception):
|
||
pass
|
||
|
||
|
||
# --------------------------------------------------------------------------- #
|
||
# Slug / paths
|
||
# --------------------------------------------------------------------------- #
|
||
|
||
|
||
def slugify(name: str) -> str:
|
||
slug = re.sub(r"[^a-z0-9_-]+", "-", name.strip().lower())
|
||
slug = re.sub(r"-{2,}", "-", slug).strip("-_")
|
||
return slug or "stack"
|
||
|
||
|
||
def stacks_root(override: Optional[str] = None) -> str:
|
||
return override or settings.STACKS_DIR
|
||
|
||
|
||
def stack_dir(stack_id: str, override: Optional[str] = None) -> str:
|
||
return os.path.join(stacks_root(override), stack_id)
|
||
|
||
|
||
def find_compose_file(directory: str) -> Optional[str]:
|
||
for name in COMPOSE_FILENAMES:
|
||
candidate = os.path.join(directory, name)
|
||
if os.path.isfile(candidate):
|
||
return candidate
|
||
return None
|
||
|
||
|
||
def compose_path(stack_id: str, override: Optional[str] = None) -> str:
|
||
directory = stack_dir(stack_id, override)
|
||
return find_compose_file(directory) or os.path.join(directory, DEFAULT_COMPOSE_NAME)
|
||
|
||
|
||
def env_path(stack_id: str, override: Optional[str] = None) -> str:
|
||
return os.path.join(stack_dir(stack_id, override), ".env")
|
||
|
||
|
||
# --------------------------------------------------------------------------- #
|
||
# Read / write files
|
||
# --------------------------------------------------------------------------- #
|
||
|
||
|
||
def read_compose(stack_id: str, override: Optional[str] = None) -> str:
|
||
path = compose_path(stack_id, override)
|
||
if not os.path.isfile(path):
|
||
return ""
|
||
with open(path, "r", encoding="utf-8") as fh:
|
||
return fh.read()
|
||
|
||
|
||
def read_env(stack_id: str, override: Optional[str] = None) -> str:
|
||
path = env_path(stack_id, override)
|
||
if not os.path.isfile(path):
|
||
return ""
|
||
with open(path, "r", encoding="utf-8") as fh:
|
||
return fh.read()
|
||
|
||
|
||
def write_compose(stack_id: str, content: str, override: Optional[str] = None) -> None:
|
||
directory = stack_dir(stack_id, override)
|
||
os.makedirs(directory, exist_ok=True)
|
||
path = compose_path(stack_id, override)
|
||
# Non-destructive: back up existing file first.
|
||
if os.path.isfile(path):
|
||
shutil.copy2(path, path + ".bak")
|
||
tmp = path + ".tmp"
|
||
with open(tmp, "w", encoding="utf-8") as fh:
|
||
fh.write(content)
|
||
os.replace(tmp, path)
|
||
|
||
|
||
def write_env(stack_id: str, content: str, override: Optional[str] = None) -> None:
|
||
directory = stack_dir(stack_id, override)
|
||
os.makedirs(directory, exist_ok=True)
|
||
path = env_path(stack_id, override)
|
||
with open(path, "w", encoding="utf-8") as fh:
|
||
fh.write(content)
|
||
|
||
|
||
def delete_stack_files(stack_id: str, override: Optional[str] = None) -> None:
|
||
directory = stack_dir(stack_id, override)
|
||
if os.path.isdir(directory):
|
||
shutil.rmtree(directory)
|
||
|
||
|
||
def clone_stack_files(src_id: str, dst_id: str, override: Optional[str] = None) -> None:
|
||
src = stack_dir(src_id, override)
|
||
dst = stack_dir(dst_id, override)
|
||
if os.path.isdir(dst):
|
||
raise StackFileError(f"Target stack '{dst_id}' already exists")
|
||
shutil.copytree(src, dst)
|
||
|
||
|
||
def discover_stacks(override: Optional[str] = None) -> list[str]:
|
||
"""Return ids of all directories under STACKS_DIR that contain a compose file."""
|
||
root = stacks_root(override)
|
||
if not os.path.isdir(root):
|
||
return []
|
||
found = []
|
||
for entry in sorted(os.listdir(root)):
|
||
# Hidden directories are StackPilot's own scratch space (a restore
|
||
# stages there before swapping it in), never a stack.
|
||
if entry.startswith("."):
|
||
continue
|
||
directory = os.path.join(root, entry)
|
||
if os.path.isdir(directory) and find_compose_file(directory):
|
||
found.append(entry)
|
||
return found
|
||
|
||
|
||
# --------------------------------------------------------------------------- #
|
||
# Live status from Docker
|
||
# --------------------------------------------------------------------------- #
|
||
|
||
|
||
@dataclass
|
||
class ContainerInfo:
|
||
id: str
|
||
name: str
|
||
service: str
|
||
image: str
|
||
state: str # running, exited, ...
|
||
status: str # human string
|
||
health: Optional[str] = None
|
||
ports: list[dict] = field(default_factory=list)
|
||
created: Optional[str] = None
|
||
|
||
|
||
def _parse_ports(attrs: dict) -> list[dict]:
|
||
ports = []
|
||
bindings = (attrs.get("NetworkSettings") or {}).get("Ports") or {}
|
||
for container_port, host in (bindings or {}).items():
|
||
if host:
|
||
for binding in host:
|
||
ports.append(
|
||
{
|
||
"container": container_port,
|
||
"host_ip": binding.get("HostIp"),
|
||
"host_port": binding.get("HostPort"),
|
||
}
|
||
)
|
||
else:
|
||
ports.append({"container": container_port, "host_port": None})
|
||
return ports
|
||
|
||
|
||
def containers_for_stack(stack_id: str) -> list[ContainerInfo]:
|
||
client = get_client()
|
||
raw = safe_call(
|
||
client.containers.list,
|
||
all=True,
|
||
filters={"label": f"{COMPOSE_LABEL}={stack_id}"},
|
||
)
|
||
result = []
|
||
for c in raw:
|
||
attrs = c.attrs
|
||
state = attrs.get("State", {}) or {}
|
||
health = (state.get("Health") or {}).get("Status")
|
||
result.append(
|
||
ContainerInfo(
|
||
id=c.id,
|
||
name=c.name,
|
||
service=c.labels.get(SERVICE_LABEL, c.name),
|
||
# Use the configured image name from the inspect we already have;
|
||
# reading c.image.tags would trigger a separate image-inspect call
|
||
# per container.
|
||
image=attrs.get("Config", {}).get("Image", "") or attrs.get("Image", ""),
|
||
state=state.get("Status", c.status),
|
||
status=attrs.get("State", {}).get("Status", c.status),
|
||
health=health,
|
||
ports=_parse_ports(attrs),
|
||
created=attrs.get("Created"),
|
||
)
|
||
)
|
||
return result
|
||
|
||
|
||
def _status_from_states(states: list[str]) -> str:
|
||
if not states:
|
||
return "stopped"
|
||
if any(s == "dead" for s in states):
|
||
return "error"
|
||
running = [s for s in states if s == "running"]
|
||
if len(running) == len(states):
|
||
return "running"
|
||
if running:
|
||
return "partial"
|
||
return "stopped"
|
||
|
||
|
||
def compute_status(stack_id: str, containers: Optional[list[ContainerInfo]] = None) -> str:
|
||
"""Status for one stack, from its containers alone.
|
||
|
||
"updating" is not derived here: whether an operation is in flight lives in
|
||
``stack_lock_service``, and callers that want to show it overlay the lock on
|
||
top of this. Pass already-fetched ``containers`` to avoid a redundant Docker
|
||
round-trip (the detail view already has them).
|
||
"""
|
||
try:
|
||
if containers is None:
|
||
containers = containers_for_stack(stack_id)
|
||
except DockerError:
|
||
return "unknown"
|
||
return _status_from_states([c.state for c in containers])
|
||
|
||
|
||
def stack_status_summaries() -> dict[str, dict]:
|
||
"""One cheap Docker call → ``{project: {status, total, running}}`` for every
|
||
stack at once.
|
||
|
||
Uses the low-level container *summary* list (``GET /containers/json``, no
|
||
per-container inspect) grouped by the compose project label. This is far
|
||
cheaper than calling :func:`containers_for_stack` (which full-inspects every
|
||
container) once per stack for the overview — turning the stacks list from
|
||
O(stacks × containers) Docker round-trips into a single one.
|
||
"""
|
||
client = get_client()
|
||
raw = safe_call(client.api.containers, all=True)
|
||
by_project: dict[str, list[str]] = {}
|
||
for c in raw:
|
||
project = (c.get("Labels") or {}).get(COMPOSE_LABEL)
|
||
if project:
|
||
by_project.setdefault(project, []).append(c.get("State", ""))
|
||
return {
|
||
project: {
|
||
"status": _status_from_states(states),
|
||
"total": len(states),
|
||
"running": sum(1 for s in states if s == "running"),
|
||
}
|
||
for project, states in by_project.items()
|
||
}
|
||
|
||
|
||
# --------------------------------------------------------------------------- #
|
||
# Compose CLI lifecycle (async subprocess)
|
||
# --------------------------------------------------------------------------- #
|
||
|
||
|
||
def _compose_base_cmd(stack_id: str, override: Optional[str] = None) -> list[str]:
|
||
directory = stack_dir(stack_id, override)
|
||
compose_file = find_compose_file(directory) or os.path.join(directory, DEFAULT_COMPOSE_NAME)
|
||
return [
|
||
"docker",
|
||
"compose",
|
||
"-p",
|
||
stack_id,
|
||
"--project-directory",
|
||
directory,
|
||
"-f",
|
||
compose_file,
|
||
]
|
||
|
||
|
||
async def run_compose(
|
||
stack_id: str,
|
||
args: list[str],
|
||
override: Optional[str] = None,
|
||
timeout: float = 600.0,
|
||
) -> dict:
|
||
"""Run a `docker compose` subcommand. Returns {returncode, stdout, stderr}."""
|
||
cmd = _compose_base_cmd(stack_id, override) + args
|
||
proc = await asyncio.create_subprocess_exec(
|
||
*cmd,
|
||
env=registry_service.cli_env(),
|
||
stdout=asyncio.subprocess.PIPE,
|
||
stderr=asyncio.subprocess.PIPE,
|
||
)
|
||
try:
|
||
stdout_b, stderr_b = await asyncio.wait_for(proc.communicate(), timeout=timeout)
|
||
except asyncio.TimeoutError as exc:
|
||
proc.kill()
|
||
raise StackFileError(f"compose command timed out: {' '.join(args)}") from exc
|
||
return {
|
||
"returncode": proc.returncode,
|
||
"stdout": stdout_b.decode("utf-8", "replace"),
|
||
"stderr": stderr_b.decode("utf-8", "replace"),
|
||
"command": " ".join(args),
|
||
}
|
||
|
||
|
||
async def validate_yaml(content: str, env_content: str = "") -> dict:
|
||
"""Validate compose YAML by running ``docker compose config -q`` against it
|
||
in a throwaway directory. Returns ``{"ok": bool, "errors": str}`` — the
|
||
errors string carries compose's own diagnostics (parse errors, unknown
|
||
keys, bad interpolation) so the editor can surface them before saving."""
|
||
with tempfile.TemporaryDirectory(prefix="sp-validate-") as tmp:
|
||
compose_file = os.path.join(tmp, DEFAULT_COMPOSE_NAME)
|
||
with open(compose_file, "w", encoding="utf-8") as fh:
|
||
fh.write(content)
|
||
if env_content:
|
||
with open(os.path.join(tmp, ".env"), "w", encoding="utf-8") as fh:
|
||
fh.write(env_content)
|
||
cmd = [
|
||
"docker", "compose",
|
||
"--project-directory", tmp,
|
||
"-f", compose_file,
|
||
"config", "-q",
|
||
]
|
||
try:
|
||
proc = await asyncio.create_subprocess_exec(
|
||
*cmd,
|
||
env=registry_service.cli_env(),
|
||
stdout=asyncio.subprocess.PIPE,
|
||
stderr=asyncio.subprocess.PIPE,
|
||
)
|
||
_out, err = await asyncio.wait_for(proc.communicate(), timeout=30.0)
|
||
except asyncio.TimeoutError as exc:
|
||
raise StackFileError("compose config validation timed out") from exc
|
||
ok = proc.returncode == 0
|
||
return {"ok": ok, "errors": "" if ok else err.decode("utf-8", "replace").strip()}
|
||
|
||
|
||
async def stream_compose(
|
||
stack_id: str, args: list[str], override: Optional[str] = None
|
||
):
|
||
"""Yield lines from a `docker compose` subcommand as they are produced."""
|
||
cmd = _compose_base_cmd(stack_id, override) + args
|
||
proc = await asyncio.create_subprocess_exec(
|
||
*cmd,
|
||
env=registry_service.cli_env(),
|
||
stdout=asyncio.subprocess.PIPE,
|
||
stderr=asyncio.subprocess.STDOUT,
|
||
)
|
||
assert proc.stdout is not None
|
||
async for raw in proc.stdout:
|
||
yield raw.decode("utf-8", "replace").rstrip("\n")
|
||
await proc.wait()
|
||
|
||
|
||
_json_progress: Optional[bool] = None
|
||
|
||
|
||
async def supports_json_progress() -> bool:
|
||
"""Whether this Docker Compose understands ``--progress json``.
|
||
|
||
The JSON progress stream carries per-layer ``current``/``total`` bytes, which
|
||
the deploy console turns into a real progress bar. Older compose releases
|
||
reject the value, so probe once (cheap, no side effects) and cache it; on a
|
||
negative result callers fall back to the plain text stream.
|
||
"""
|
||
global _json_progress
|
||
if _json_progress is None:
|
||
try:
|
||
proc = await asyncio.create_subprocess_exec(
|
||
"docker", "compose", "--progress", "json", "version",
|
||
env=registry_service.cli_env(),
|
||
stdout=asyncio.subprocess.DEVNULL,
|
||
stderr=asyncio.subprocess.DEVNULL,
|
||
)
|
||
rc = await asyncio.wait_for(proc.wait(), timeout=15.0)
|
||
_json_progress = rc == 0
|
||
except Exception: # noqa: BLE001 - probe failure just disables the feature
|
||
_json_progress = False
|
||
return _json_progress
|
||
|
||
|
||
async def _stream_phase(
|
||
stack_id: str, args: list[str], override: Optional[str], json_progress: bool
|
||
):
|
||
"""One compose subcommand, streamed. Yields ``("log", line)`` per output
|
||
line, then ``("rc", returncode)`` exactly once."""
|
||
cmd = _compose_base_cmd(stack_id, override)
|
||
if json_progress:
|
||
# Global flag, must precede the subcommand.
|
||
cmd += ["--progress", "json"]
|
||
cmd += args
|
||
proc = await asyncio.create_subprocess_exec(
|
||
*cmd,
|
||
env=registry_service.cli_env(),
|
||
stdout=asyncio.subprocess.PIPE,
|
||
stderr=asyncio.subprocess.STDOUT,
|
||
)
|
||
assert proc.stdout is not None
|
||
async for raw in proc.stdout:
|
||
yield ("log", raw.decode("utf-8", "replace").rstrip("\n"))
|
||
await proc.wait()
|
||
yield ("rc", proc.returncode)
|
||
|
||
|
||
async def stream_up(stack_id: str, override: Optional[str] = None):
|
||
"""Run `compose up -d` streaming combined output, so the deploy console can
|
||
show image-pull and container-create progress live.
|
||
|
||
Yields ``("log", line)`` for each output line, then ``("done", returncode)``.
|
||
"""
|
||
json_progress = await supports_json_progress()
|
||
async for kind, payload in _stream_phase(
|
||
stack_id, ["up", "-d", "--remove-orphans"], override, json_progress
|
||
):
|
||
yield ("done", payload) if kind == "rc" else ("log", payload)
|
||
|
||
|
||
async def stream_update(stack_id: str, override: Optional[str] = None):
|
||
"""Run `compose pull` then `compose up -d`, streaming both phases, so the
|
||
stacks list can show real update progress instead of a spinner.
|
||
|
||
Yields ``("log", line)`` for each output line of either phase, then
|
||
``("done", returncode)`` once. A failed pull short-circuits: recreating
|
||
containers on images that never came down would only make things worse.
|
||
"""
|
||
json_progress = await supports_json_progress()
|
||
rc = 0
|
||
for args in (["pull"], ["up", "-d", "--remove-orphans"]):
|
||
async for kind, payload in _stream_phase(stack_id, args, override, json_progress):
|
||
if kind == "log":
|
||
yield ("log", payload)
|
||
else:
|
||
rc = payload
|
||
if rc != 0:
|
||
break
|
||
yield ("done", rc)
|
||
|
||
|
||
# Convenience lifecycle wrappers ------------------------------------------------
|
||
|
||
|
||
async def up(stack_id: str, override: Optional[str] = None) -> dict:
|
||
return await run_compose(stack_id, ["up", "-d", "--remove-orphans"], override)
|
||
|
||
|
||
async def down(stack_id: str, override: Optional[str] = None) -> dict:
|
||
return await run_compose(stack_id, ["down"], override)
|
||
|
||
|
||
async def start(stack_id: str, override: Optional[str] = None) -> dict:
|
||
return await run_compose(stack_id, ["start"], override)
|
||
|
||
|
||
async def stop(stack_id: str, override: Optional[str] = None) -> dict:
|
||
return await run_compose(stack_id, ["stop"], override)
|
||
|
||
|
||
async def restart(stack_id: str, override: Optional[str] = None) -> dict:
|
||
return await run_compose(stack_id, ["restart"], override)
|
||
|
||
|
||
async def pull(stack_id: str, override: Optional[str] = None) -> dict:
|
||
return await run_compose(stack_id, ["pull"], override)
|
||
|
||
|
||
async def update(stack_id: str, override: Optional[str] = None) -> dict:
|
||
"""Pull then up -d."""
|
||
pull_res = await run_compose(stack_id, ["pull"], override)
|
||
up_res = await run_compose(stack_id, ["up", "-d", "--remove-orphans"], override)
|
||
return {
|
||
"returncode": up_res["returncode"],
|
||
"stdout": pull_res["stdout"] + "\n" + up_res["stdout"],
|
||
"stderr": pull_res["stderr"] + "\n" + up_res["stderr"],
|
||
"command": "pull + up -d",
|
||
}
|
||
|
||
|
||
async def logs(
|
||
stack_id: str,
|
||
service: Optional[str] = None,
|
||
tail: int = 200,
|
||
override: Optional[str] = None,
|
||
) -> dict:
|
||
args = ["logs", "--no-color", "--tail", str(tail), "--timestamps"]
|
||
if service:
|
||
args.append(service)
|
||
return await run_compose(stack_id, args, override, timeout=60.0)
|
||
|
||
|
||
def now() -> datetime:
|
||
return datetime.now(timezone.utc)
|