Skip to content
LabTether

Changes in 2.0

LabTether 2.0 has a new web console. Every page has a new layout, and some pages moved. Page addresses are the same, so bookmarks keep working, and your theme and language carry over. The hub's API, database, agents, connectors and settings work as they did in v1.7.1, so an upgrade needs no configuration change. Two exceptions: the audit log now keeps changes only, not every API request, and the audit list in the API gives non-admin users and keys only their own rows (see Changed). To upgrade, see Upgrading.

Where pages moved

The sidebar now lists Dashboard, Devices, Map, Services, Containers, Alerts, Automations and Logs, then Terminal, Files and Remote View, then Settings and your account. Below 900 px wide it folds into a menu. The account menu has theme, language and Sign out. The v1.7.1 console had no way to sign out.

In v1.7.1 In 2.0
Topology Map
Incidents Alerts → Issues. Each issue has its own page.
Updates Devices → ⋯ → Updates
Actions, Schedules, Webhooks Automations, one tab each
Users, Groups, Health, Audit Log Settings, one tab each
Recordings Settings → Audit log → Recordings
Settings → General: notification channels Settings → Notifications
Settings → General: Prometheus Settings → Security
Settings → API access: API keys, MCP, API reference Settings → Security
Settings → Advanced: credential profiles Settings → Security → Saved logins
Settings → Advanced: SSH hub key Settings → Agents
Settings → Advanced: retention, runtime settings, service discovery defaults Settings → General
Security: connect agents and enrollment tokens Settings → Agents
Security: database, reset Settings → General
Security: TLS, Tailscale Serve, single sign-on Settings → Security

Old links to /settings?tab=... open the section that holds those settings now.

Settings tabs

Tab What it holds Who can open it
General Appearance, retention, advanced runtime settings, service discovery defaults, database, config export, version and reset Admin and owner
Notifications Channels: email, Slack, ntfy, Gotify and webhooks, each with Send test Admin and owner
Users Accounts and roles, and your own password and two-factor sign-in Everyone
Groups The group tree: each group's name, parent (Inside) and the jump hosts to reach it (Reach through) Everyone
Security Certificate, Tailscale Serve, single sign-on (OIDC), saved logins, API keys, Prometheus, MCP and the API reference Admin and owner
Agents The hub address agents connect to, enrollment tokens, agent keys and the hub SSH key Admin and owner
Health The hub's own checks, downtime windows (Schedule downtime), each group's health score and events, and failed background jobs Everyone
Audit log Changes made by people and API keys, newest first, 200 at a time with Load older entries, refused ones marked Denied, Export CSV, and Recordings as a second view. The owner and admins see everyone's entries; others see their own Everyone; Recordings need the operator role or above

Pages

  • Dashboard: Needs attention, Running, Servers and Services, and a small map, with links to Alerts, Devices, Services and the Map.
  • Devices: one list in sections by type (Servers, Storage, Computers and VMs), with CPU, memory and disk bars. Devices that need attention come first, and VMs fold under their Proxmox host. Filter by name, IP, OS or tag, or group the list by type, status or group. The group tree with drag and drop is under ⋯ → Arrange groups.
  • Device page: a new header, a tab for each thing it supports (for example Overview, System, Logs, Proxmox, Docker, TrueNAS, Terminal, Files, Remote view and Settings) and a ⋯ menu with Terminal, Files, Remote view, Copy address, Edit, Add connection, Restart, Shut down and Remove. The same menu is on each row of the Devices list.
  • Map: a top-down tree of your network built from the hub's topology data. Select a device to see its address, OS, what it runs on, what it contains and its CPU over 24 hours. See Map.
  • Alerts: the alert list beside the selected alert, with Acknowledge, then Resolve, as the main button. Tabs: Alerts, Issues, Rules, Silences and Delivery log. Rules can be turned on or off and edited in place. See Alerts.
  • Issues: each issue has its own page with its state, owner, timeline, notes and a write-up with root cause, follow-ups and lessons learned. See Issues.
  • Updates: update plans and their run history. See Updates.
  • Containers: one list of every container, grouped by host or stack, with Running and Stopped filters, CPU and memory, and a ⋯ menu with Logs, Shell, Restart, Start and Stop. Run container and Deploy stack open forms. Docker and Portainer hosts have their own tab on the device page.
  • Terminal: a start page with open sessions, recent devices and all devices.
  • Files: a tab per device or connection, a path bar, a right-click menu, a second pane and one bar for all transfers.
  • Remote View: a start page with devices and bookmarks, and a toolbar that hides in full screen. The viewer protocols are unchanged.
  • Terminal, Files and Remote view are also tabs on each device page.
  • Command palette: Cmd+K or Ctrl+K opens pages, devices and quick connect. See Keyboard shortcuts.

Changed

  • The audit log keeps a row when someone changes something, or tries to and is refused, instead of one for every API request. Reads go to the hub's log only. Read rows stored by older hubs stay in the database until audit retention removes them, but the list no longer shows them.
  • GET /audit/events, GET /api/v2/audit/events and labtether-cli audit give the owner, admins and admin API keys every row. Other users and keys get only the rows they caused.
  • Every time in the console uses the 24-hour clock.
  • The old Map layout editor is under Map → ⋯ → Edit layout.

Removed

From the Dashboard: the summary paragraph, the number cards, the export button, and the Fleet Focus, Activity, Resource Hotspots and topology panels. Containers in trouble show under Needs attention.

Fixed

  • Settings → Security → Saved logins lists your saved logins, and the login pickers for terminals, remote view and file connections offer them. In v1.7.1 the list did not load on a real hub.
  • Saved webhooks now show in the Webhooks list. In v1.7.1 the list stayed empty.
  • Container links on hosts with a dash in their name open the right container.
  • In the command palette, Enter on a plain word no longer opens a terminal.
  • /api-docs opens the API reference without a language prefix in the address.
  • Clearing a schedule's group when you edit it is saved.
  • Terminal links with ?target= or ?quickConnect= open that device.
  • Reconnecting to a terminal no longer types stray text such as 0c1;2c0;276 at the prompt.
  • A terminal whose connection cannot open shows the error with Try again, instead of retrying several times a second. If the stream gets no answer at all, the error shows after 45 seconds.
  • The Linux install command in Add device asks for the enrollment token and uses the token you paste. In v1.7.1 it read the command's next line as the token, so the agent could not enroll, and the pasted token ran as a command in your shell.
  • When the hub's status or alert list does not load, Devices, the Dashboard and Alerts say so, with Try again. In v1.7.1 they showed "No devices yet" or "No alerts".
  • An API key past its expiry date shows Expired.
  • The Audit log shows people's actions again. In v1.7.1 every open console added an "API request GET" row every few seconds.
  • Percent charts no longer show an axis above 100% on a busy device.
  • Settings → Security → Prometheus shows https://<hub-host>:8443/metrics as the scrape address when hub TLS is on.
  • Changing a connection's type on the Map keeps its label.
  • An agent at the hub's agent version or newer shows as up to date. In v1.7.1 any other version showed an update, so an agent updated before the hub was offered the older release, which the agent refuses.