Subtrack Railway Deploy

SubTrack Railway Deploy

Scope

Operate the SubTrack Telegram bot deployment.

• Local repo: /Users/ris/Documents/vibecoding_11
• GitHub repo: https://github.com/serejaris/vibecoding_11
• Railway project: brilliant-motivation
• Railway environment: production
• Railway service: vibecoding_11
• Telegram bot: @ris_subtrack_bot
• Start command: python main.py
• Runtime mode: long polling, exactly one active process per token

Do not print, paste, commit, or summarize real token values.

First Response

State the current surface: local repo, GitHub, Railway, or Telegram runtime. If the user reports a crash, inspect the logs first. If the screenshot or logs show BOT_TOKEN не найден, fix Railway variables before changing code.

Quick Check

Run the bundled script for a safe read-only snapshot:

/Users/ris/.codex/skills/subtrack-railway-deploy/scripts/check_subtrack_deploy.sh

The script checks local repo status, required files, local env presence without printing secrets, Railway link/status, recent deployments, recent logs, Telegram getMe, webhook state, and local process conflicts.

Deploy Or Repair Workflow

1. Work from /Users/ris/Documents/vibecoding_11.
2. Check local status and avoid touching unrelated dirty files.
3. Verify tests when code changed:

.venv/bin/python -m unittest discover -s tests -v
git diff --check

4. Ensure Railway CLI is installed and linked:

railway link -p brilliant-motivation -e production -s vibecoding_11
railway status

5. If Railway lacks variables, copy from local .env without printing values:

set -a
. ./.env
set +a
railway variables --set "BOT_TOKEN=$BOT_TOKEN" --set "REMIND_DAYS_BEFORE=${REMIND_DAYS_BEFORE:-3}" --skip-deploys

6. Redeploy:

railway redeploy -s vibecoding_11 --yes
railway deployment list -s vibecoding_11 -e production

7. Inspect logs. A healthy startup contains:

Start polling
Run polling for bot @ris_subtrack_bot

8. If logs contain Telegram conflict, stop the other process using the same token. Check local first:

ps aux | rg -i 'vibecoding_11|main.py|aiogram' | rg -v rg

Kill only the SubTrack local process after confirming it is the duplicate. Do not stop unrelated Python bots.

9. Verify Telegram token identity without exposing it:

set -a
. ./.env
set +a
python3 - <<'PY'
import os, json, urllib.request
url = f"https://api.telegram.org/bot{os.environ['BOT_TOKEN']}/getMe"
with urllib.request.urlopen(url, timeout=10) as r:
    data = json.load(r)
print({"ok": data.get("ok"), "username": data.get("result", {}).get("username"), "id": data.get("result", {}).get("id")})
PY

Common Failures

• RuntimeError: BOT_TOKEN не найден: Railway env var is missing. Add BOT_TOKEN in Railway variables, then redeploy.
• TelegramConflictError: Conflict: terminated by other getUpdates request: the same token is running in another polling process. Stop local python main.py or the older cloud deployment.
• Empty railway logs --lines: use railway deployment list to get the latest deployment id, then call railway logs <deployment-id> --deployment --lines 120.
• Local tests fail with No module named pytest: this repo uses unittest; run .venv/bin/python -m unittest discover -s tests -v.

Safety Rules

• Do not commit .env, data/*.db, logs, or token values.
• Do not rotate the Telegram token unless the user explicitly asks.
• Do not claim production is fixed until Railway logs show polling or Telegram runtime is otherwise verified.
• Keep SQLite unless the user asks for production storage changes.