Skip to content
Tickwise

Telegram bot

Network failures — in one Telegram message

What the Tickwise bot does: incident cards with cause and steps, a network status menu, notification settings and daily digests.

On this page

In the closed beta Tickwise has no web dashboard: everything is managed through the Telegram bot. It sends server crash alerts to Telegram, shows the state of your network and servers and the list of incidents, and lets you choose what gets sent and when. Almost everything is done with buttons: the bot edits a single message as you move between screens, and every screen has “Back” and “Menu”.

The bot only manages notifications and incidents. It never sends commands to your servers — the agent cannot run them.

Getting connected

The bot starts working in a chat once the first server is linked: run tickwise link in the console, then send /link CODE to the bot. The full walkthrough is in the installation guide.

  • You can link a server in a private chat with the bot or in a group. In a group, /link is accepted only from chat administrators.
  • The chat where the first server was linked becomes the organisation. /link in another chat creates a separate organisation.
  • In a chat that is not linked, every command shows short linking instructions.

/menu opens the main screen: for each network, how many servers are online, how many incidents are open and the worst TPS over the last 5 minutes.

🏠 My network

🌐 Network default

🟡 online 3/4 · 🚨 open incidents: 1

📉 worst TPS over 5 min: 17.8 (survival)

Updated 10:19 MSK

10:19

🌐 Servers (4)🚨 Incidents (1)
🔔 Notifications📊 Digests
⚙️ Manage❓ Help
🔄 Refresh
The main screen (translated). Buttons open sections in the same message.
CommandOpens
/menu (also /start, /status)the main screen
/serversthe server list
/incidentsopen incidents
/settingsthis chat's notification settings
/digestdigests
/link CODE [name] [network]link a server
/helphelp
/privacywhat data Tickwise receives
/cancelcancel typing a new name when renaming

Servers

The list shows every server with an indicator: 🟢 healthy, 🟡 degraded, 🔴 down, ⚫ agent silent for more than 90 seconds, 🛠 maintenance, ⚪ the agent has never reported in. Muted servers are marked 🔕.

A server card shows:

  • the role (server, backend or proxy), platform and version, Java version, agent and signature pack versions, and when the agent last reported in;
  • for the last 5 minutes — TPS and MSPT (average and p95), player count and memory in use;
  • for the last 24 hours — the worst TPS and peak MSPT (for backends and standalone servers);
  • the number of plugins and buttons for this server's open incidents.

Card buttons:

  • 🔕 1 h / 🔕 8 h / 🔕 24 h — mute this server's notifications in this chat. To undo early, press “Restore notifications”.
  • ✏️ Rename — the bot asks you to reply with the new name within 5 minutes: 1–32 characters, letters, digits, space, _ . -, unique within the network. The old name is kept as an alias for matching against velocity.toml.
  • 🔌 Unlink — after confirmation, revokes the agent's token: the server stops sending data and its notifications stop, while local diagnostics on the server keep working. To connect it again, run tickwise link in the console.

Incidents

/incidents opens a list with “Open” and “Last 7 days” tabs, 5 incidents per page, paged with ◀ ▶. When a new incident opens, the bot sends its card as a separate message; the list opens the same card.

🔴 CRITICAL · network default

Backend “lobby” is down: server crashed — players can't join

Servers: velocity (proxy), lobby (probable root cause)

Cause: the proxy cannot connect players to lobby. Probable root cause — a crash on the backend itself (Exception in server tick loop).

What changed before the failure (a suspicion, not proof):

• lobby: LuckPerms 5.4.140 → 5.4.141

🕒 02.10 07:19 UTC

10:19

🔎 Details
👍 Helped👎 Didn't help
✅ On it✔️ Resolved
An incident card (translated): cause, confidence, evidence and steps.

The card shows the severity, the network, the servers and the probable root cause, the cause and how reliable it is (for example a verified fix, a probable cause or a hypothesis), up to three pieces of evidence, changes made before the failure and up to three “what to do” steps.

  • 🔎 Details opens the incident page through a secret link valid for 30 days. Every time the card is opened from the menu, a fresh link is created.
  • 👍 Helped / 👎 Didn't help feed into the signature's statistics. 👎 on a “new error” card mutes that error on the server: you will not be alerted about it again.
  • ✅ On it and ✔️ Resolved change the incident's status. A closed incident has no such buttons.

If the cloud refines the root cause, the bot edits the card it already sent (marked “🔄 Updated”) instead of sending a new message.

Recovery messages

When an incident closes on its own — the failure signals have stopped — the bot edits the card (“✅ Recovered · lasted 7 min”) and replies to it with a short “✅ Recovered” message so that you get a notification. This applies to incidents whose card was already sent to the chat. You can turn it off in the notification settings.

Notification settings

/settings configures notifications for the current chat.

🔔 Notifications for this chat

Notifications: ✅ on

Severity: all

Quiet hours: 23:00–08:00 MSK — critical only

Report recovery: ✅

10:19

🔕 Turn notifications off
• Allerror+critical
No quiet hours• 23–0800–09
• MSKUTC
✅ Report recovery
← Back🏠 Menu
The settings screen (translated).
SettingOptionsDefault
Notificationson or offon
Severityall, error and above, critical onlyall
Quiet hoursoff, 23–08, 00–09off
Time zoneMoscow (MSK) or UTCMoscow
Report recoveryyes or noyes

During quiet hours only critical incidents come through, while recovery messages and digests arrive silently. If every server involved in an incident is muted in this chat, the card is not sent. Muted servers are listed on the same screen, each with a button to restore it.

Digests

/digest sends a digest on demand — for the last day or the last week — and sets a schedule: off (the default), daily at 10:00, or Mondays at 10:00 in the chat's time zone. The first scheduled digest arrives at the next 10:00, not immediately.

A digest includes:

  • how many incidents were opened (and how many of them critical), closed and are open now;
  • the three most frequent categories;
  • the number of new unknown errors;
  • for each server — the share of time it was online, the worst TPS and peak MSPT.

Management and group permissions

The “⚙️ Manage” section shows the organisation, the number of servers and the linked chats.

  • ✏️ Rename organisation — up to 48 characters.
  • 🚪 Unlink this chat — notifications stop coming here; servers and history stay. If this is the organisation's only chat and servers are still linked, the bot asks you to unlink the servers first.

In a private chat its owner can change everything. In a group, every member can view the screens, but only chat administrators can change settings, mute, rename and unlink. Everyone else gets a reply that only chat administrators can change settings in a group.

What exactly Tickwise receives from your servers is described on the Security page.