AssRep ops repo for host monitoring, log alerts, and docker cleanup.
- Host metrics alerts (CPU/RAM/DISK) via Telegram.
- Docker prune with "before/after" notification.
- Log-alert container (regex-based error scanning).
- Ops bot with slash commands for restart/prune/status.
- OPS_SSH_HOST, OPS_SSH_USER, OPS_SSH_KEY, OPS_SSH_PORT
- OPS_TG_TOKEN, OPS_TG_CHAT_ID
- OPS_LOG_ALERT_REGEX
- OPS_ALLOWED_CHATS
- OPS_COMPOSE_PROJECT
- OPS_SERVICES
- OPS_API_HEALTH_URL
- OPS_ENABLE_HOST_MONITOR (default 1)
- OPS_ENABLE_DOCKER_PRUNE (default 1)
- OPS_ENABLE_LOG_ALERT (default 0)
- OPS_ENABLE_OPS_BOT (default 1)
- Optional: OPS_CPU_LIMIT, OPS_RAM_LIMIT, OPS_DISK_LIMIT, OPS_SEND_OK
- Optional: OPS_ALERT_REMINDER_SEC, OPS_OK_REMINDER_SEC
- Optional: OPS_LOG_ALERT_* (project/env/include/exclude/window/flush/top_n/sample_len)
- Optional: OPS_LOG_ALERT_IMAGE
Host monitor anti-spam:
- sends immediately on entering alert state
- sends again only when alert set changes or reminder cooldown passes (
OPS_ALERT_REMINDER_SEC) - sends one recovery message when metrics return to normal
- optional OK heartbeat (
OPS_SEND_OK=1) is rate-limited byOPS_OK_REMINDER_SEC
ansible-playbook ansible/playbook.yml -i inventory.ini
- Host monitor uses /etc/assrep/monitor.env
- Log alert uses /etc/assrep/log-alert.env
- Ops bot uses /etc/assrep/ops-bot.env
- /menu (inline buttons)
- /status
- /containers
- /restart
- /restart_all
- /logs [lines]
- /prune
- /ping
- /mute
- /unmute
Ops bot UI:
- inline action menu with status/containers/ping/restart/logs/prune
- confirmation prompts for restart_all/prune
- quick mute buttons (30m/1h/3h) from menu and directly in alert messages