Server Groups carve the fleet into meaningful buckets — production, staging, web-tier, db-cluster, client-acme. Once a server sits in a group, that group becomes a first-class filter across the platform.

What you see

At /server-groups you get one row per group with:

• Group name and colour dot.
• Number of servers in the group.
• Aggregated online / offline counts.
• Total monthly cost (summed from each member’s cost).
• Last edited timestamp.

Open a group and you get its member list, a mini KPI strip for that subset, and the same three-dot server actions available from the main Servers page.

What you can do

• New Group — opens a two-step modal:
  1. Details — name, description, colour.
  2. Servers — pick members from a searchable list.
• Edit — rename, re-colour, add / remove members.
• Delete — removes the group. Member servers are untouched.

Where groups show up

Once a group exists, it becomes a filter and a targeting primitive:

• Servers — filter dropdown narrows the fleet to one group.
• Alert rules — target “any server in group X” rather than picking hosts individually.
• Recipes — run a script against every server in a group.
• Reports — cost roll-ups are broken down per group.
• Dashboards — group filter on capacity, uptime, and heartbeat widgets.

How it works

Groups are pure metadata. Assigning a server to a group does not change anything on the host — no tags are pushed, no agent config is rewritten. Removing a server from a group only detaches it from the label.

A server can belong to more than one group. Cost roll-ups avoid double-counting a shared server across groups on the totals page.

Related

• Servers — where group filters get used most.
• Alerts & Incidents — target rules at whole groups.
• Recipes — run scripts fleet-wide, scoped by group.