Files
computer-playbook/cli/administration/deploy/development/env.py
T
kevinveenbirkenbachandClaude Opus 5 4d09286d8b fix(compose): reserve a GPU only where the daemon has one
A device reservation is not conditional in compose, and a count of 0 does not disable it: the daemon still selects the nvidia driver and fails where there is none. Every GPU-less host therefore lost compose up with 'could not select device driver', which is what broke test-dns on all five CI distros.

Move the reservation out of compose.yml and i18n/compose.override.yml into compose/gpu.override.yml, layered by compose_file_args() only when Profile.gpu_reservable() sees a non-zero INFINITO_GPU_COUNT. A GPU host renders exactly what it rendered before, so the passthrough sys-svc-container needs to find /dev/nvidia0 and emit runtime: nvidia for a gpu:true service stays intact.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-25 09:47:50 +02:00

51 lines
1.8 KiB
Python

from __future__ import annotations
import os
from utils.distros import distro_names
VALID_DISTROS: tuple[str, ...] = distro_names()
def compose_file_args() -> list[str]:
"""Compose `-f` flags shared by up and down flows.
Each override is gated on the resource it needs being present, not on the
instance slot: an override whose `:?` guard is unsatisfied would abort the
stack outright, so a checkout that was never handed a shared git dir or
cache network simply does not layer that file in.
"""
from .profile import Profile
profile = Profile()
out = ["-f", "compose.yml"]
if profile.shared_git_dir():
out += ["-f", "compose/worktree.override.yml"]
if profile.cache_stack_enabled():
out += ["-f", "compose/cache.override.yml"]
if profile.shared_cache_network():
out += ["-f", "compose/cache.shared.override.yml"]
if profile.gpu_reservable():
out += ["-f", "compose/gpu.override.yml"]
if (os.environ.get("INFINITO_PUBLISH_PORTS") or "").strip().lower() == "false":
out += ["-f", "compose/noports.override.yml"]
out += ["-f", "i18n/compose.override.yml"]
return out
def resolve_distro() -> str:
"""Return INFINITO_DISTRO; raise SystemExit if missing or invalid."""
distro = os.environ["INFINITO_DISTRO"].strip()
if not distro:
raise SystemExit(
"INFINITO_DISTRO is not set. Run 'make dotenv' (or source scripts/meta/env/load.sh) "
f"or export INFINITO_DISTRO=<{'|'.join(VALID_DISTROS)}> "
"before invoking cli.administration.deploy.development."
)
if distro not in VALID_DISTROS:
raise SystemExit(
f"INFINITO_DISTRO={distro!r} is not a valid distro. "
f"Valid: {', '.join(VALID_DISTROS)}."
)
return distro