fix: make bot startup resilient (self-healing state, gated cogs, loud failures)

Container/native startup died silently on any missing file/service. Now:

constants.py - self-healing runtime layout at import:
- create missing dirs (log dir, LOGSTORE, GRAPHICS_PATH, MUSIC_FOLDER)
- seed missing state files from the repo templates shipped next to
  constants.py (settings/system_gpt/pamiec/pamiec_muzyki/accident_log),
  falling back to safe empty JSON; existing files are NEVER overwritten

bot.py:
- log to stdout too, so 'docker logs' finally shows the crash reason
- missing Discord token = loud sys.exit with mount/env instructions
  (was: silent return -> container crash-loop with empty logs)
- every cog loads independently (one broken cog = skipped with traceback,
  bot continues)
- musician/librarian cogs are health-gated: enabled only when the service
  answers HTTP; a watchdog re-checks every 5 min and enables them the
  moment the service comes alive (no restart needed); tree re-synced
- on_ready reconnects no longer re-load extensions

requirements_conan.txt + Dockerfile.bot: aiomcrcon (Python <=3.11 only)
moved to best-effort extras so the 3.13 image builds clean and the
conanjurer cog stays dormant without it.

DOCKER_PROXMOX.md: startup model (core vs gated cogs) + crash-loop
troubleshooting incl. the 'disappearing files' checklist (nothing in the
stack deletes host files; bind mount = live state).

Verified: fresh-volume seeding creates dirs+templates, existing files
untouched, missing-token exits with FATAL message, health-gating logic
(stub-based runpy tests).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Michal Tuszowski
2026-07-08 00:08:31 +02:00
committed by Michał Tuszowski
parent 759c04a48f
commit 020a6b114a
5 changed files with 266 additions and 26 deletions
+154 -23
View File
@@ -4,9 +4,21 @@
# pylint: disable=too-many-lines
"""
Module of a python bot named Conjurer - used to work on BDSM discord servers.
Startup is defensive by design:
* every cog is loaded independently - one broken/missing dependency disables
that cog only, never the whole bot,
* cogs that need a sibling service (musician / librarian) are only enabled
after a positive health check; a watchdog keeps re-checking and enables them
the moment the service comes alive,
* fatal misconfiguration (missing Discord token) exits loudly on stderr with a
clear message instead of dying silently into a log file.
"""
import asyncio
import logging
import os
import sys
# *=========================================== Standard Library Imports
import random
@@ -15,12 +27,26 @@ from logging import handlers
# *==============Imported libraries
import discord
import requests
from discord.ext import commands
from communication_subroutine import comm_subroutine
from constants import ENCODING, LOGFILE, TOKEN
from constants import (
ENCODING,
FILE_SERVICE_ADDRESS,
GET_MP3,
LIBRARIAN_SERVICE_ADDRESS,
LOGFILE,
TOKEN,
)
logger = logging.getLogger("discord")
logger.setLevel(logging.DEBUG)
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
# File log (rotated). constants ensures the directory exists; this is belt and
# braces for exotic overrides.
os.makedirs(os.path.dirname(LOGFILE) or ".", exist_ok=True)
handler = handlers.RotatingFileHandler(
filename=LOGFILE,
encoding=ENCODING,
@@ -28,10 +54,15 @@ handler = handlers.RotatingFileHandler(
maxBytes=6 * 1024 * 1024,
backupCount=6,
)
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
handler.setFormatter(formatter)
logger.addHandler(handler)
# Console log so `docker logs` / journalctl actually show what happened.
console_handler = logging.StreamHandler()
console_handler.setLevel(logging.INFO)
console_handler.setFormatter(formatter)
logger.addHandler(console_handler)
# *=========================================== Initializations
intents = discord.Intents.default()
intents.message_content = True
@@ -48,37 +79,129 @@ intents.moderation = True
random.seed()
client = commands.Bot(intents=intents, command_prefix="$")
# *=========================================== Extension groups
# Core cogs depend on nothing but the bot itself (broken ones are skipped
# individually). Service groups are gated on a health check of the service
# they talk to and enabled later by the watchdog when the service appears.
CORE_EXTENSIONS = [
"administration_commands",
"ai_commands",
"other_commands",
"latex_commands",
"voice_recognition_commands",
"conanjurer_commands",
]
SERVICE_EXTENSION_GROUPS = {
# musician (file service): music download/search, radio control, file shares
"musician": {
"health_url": f"{FILE_SERVICE_ADDRESS}{GET_MP3}",
"extensions": ["music_commands", "radio_commands", "file_search_commands"],
},
# librarian: DOI / Crossref search
"librarian": {
"health_url": f"{LIBRARIAN_SERVICE_ADDRESS}/",
"extensions": ["librarian_commands"],
},
}
SERVICE_RECHECK_SECONDS = 300
def _service_alive(url: str) -> bool:
"""True when the service answers HTTP at all (any status code counts)."""
try:
requests.get(url, timeout=3)
return True
except requests.exceptions.RequestException:
return False
async def _load_extension_safe(name: str) -> bool:
"""Load one extension; log and continue instead of killing startup."""
if name in client.extensions:
return True
try:
await client.load_extension(name)
logger.info("Extension loaded: %s", name)
return True
except Exception: # pylint: disable=broad-exception-caught
logger.exception("Extension FAILED (disabled, bot continues): %s", name)
return False
async def _load_service_groups() -> bool:
"""Health-check each service group and load its cogs when alive.
Returns True when anything new was loaded (caller may want to re-sync).
"""
loaded_any = False
for service, group in SERVICE_EXTENSION_GROUPS.items():
missing = [e for e in group["extensions"] if e not in client.extensions]
if not missing:
continue
alive = await asyncio.to_thread(_service_alive, group["health_url"])
if not alive:
logger.warning(
"Service '%s' unreachable (%s) - cogs stay disabled: %s",
service,
group["health_url"],
", ".join(missing),
)
continue
logger.info("Service '%s' is alive - enabling: %s", service, ", ".join(missing))
for extension in missing:
if await _load_extension_safe(extension):
loaded_any = True
return loaded_any
async def _sync_tree() -> None:
try:
await client.tree.sync()
except Exception: # pylint: disable=broad-exception-caught
logger.exception("Slash-command tree sync failed (commands may lag)")
async def _service_watchdog() -> None:
"""Periodically retry offline services and enable their cogs when up."""
while not client.is_closed():
await asyncio.sleep(SERVICE_RECHECK_SECONDS)
try:
if await _load_service_groups():
await _sync_tree()
except Exception: # pylint: disable=broad-exception-caught
logger.exception("Service watchdog tick failed")
_STARTUP_DONE = False
# *=========================================== Define Events
@client.event
async def on_ready():
"""Metoda wywoływana przy połączeniu do serwera."""
logger = logging.getLogger("discord")
logger.debug("SAMPLE DEBUG LOG")
global _STARTUP_DONE # pylint: disable=global-statement
logger.info("%s has connected to Discord!", client.user)
# TODO: load vs reload
if _STARTUP_DONE:
logger.info("Reconnected - extensions already loaded")
return
_STARTUP_DONE = True
logger.info("Reactor: online")
await client.load_extension("administration_commands")
for extension in CORE_EXTENSIONS:
await _load_extension_safe(extension)
await client.load_extension("librarian_commands")
await client.load_extension("music_commands")
await client.load_extension("radio_commands")
await client.load_extension("ai_commands")
await client.load_extension("other_commands")
await client.load_extension("voice_recognition_commands")
await client.load_extension("file_search_commands")
await client.load_extension("latex_commands")
await client.load_extension("conanjurer_commands")
await _load_service_groups()
logger.info("Sensors: online")
logger.info(client.cogs)
await client.tree.sync()
await _sync_tree()
for com in client.commands:
logger.info("Command %s is awejleble", com.qualified_name)
asyncio.create_task(_service_watchdog())
logger.info("Logged in as ----> %s", client.user)
logger.info("ID:%s ", client.user.id)
logger.info("All systems: operational")
@@ -106,15 +229,10 @@ async def _run_bot(token: str, shutdown_event: asyncio.Event) -> None:
async def main() -> None:
if not TOKEN:
logger.error("Discord token missing - set DISCORD_TOKEN or configure netrc")
return
logger.info("Starting discord bot")
shutdown_event = asyncio.Event()
comm_stop_event = threading.Event()
comm_task = asyncio.create_task(_run_comm_subroutine(comm_stop_event))
bot_task = asyncio.create_task(_run_bot(TOKEN, shutdown_event))
@@ -133,4 +251,17 @@ async def main() -> None:
# *================================== Run
if __name__ == "__main__":
if not TOKEN:
# Loud, unmissable and in `docker logs`: this is THE most common cause
# of a silent container crash-loop.
MSG = (
"FATAL: Discord token missing.\n"
"Provide it via the DISCORD_TOKEN environment variable or a netrc "
"file (machine 'discord') at the path in CONJURER_NETRC_FILE.\n"
"Docker: check that your secrets mount exists, e.g.\n"
" -v /srv/conjurer/secrets/.netrc:/secrets/.netrc:ro\n"
" CONJURER_NETRC_FILE=/secrets/.netrc"
)
logger.critical(MSG)
sys.exit(MSG)
asyncio.run(main())