Group + Instance Commands

A group is the scheduling unit (think Deployment); an instance is a single Minecraft server process scheduled into a group. These commands cover the CRUD on groups, day-to-day instance ops, and the crash-report viewer for post-mortems.

What you’ll learn

How to declare a group with scaling, routing, and template layers.
How to start, stop, exec into, and tail the console of an instance.
How to find the most recent crashes and inspect their tail logs.

`prexorctl group`

`group list`

prexorctl group list
prexorctl group list --filter lobby --sort players
prexorctl group list --watch

Flags:

--filter <substr> — substring match against group name.
--sort <key> — name (default), players, or instances.
--watch — re-render every 2s; press Ctrl-C to exit.

`group info`

prexorctl group info <name>

Shows three side-by-side cards (config, scaling, template) plus the running-instance table for the group.

`group create`

prexorctl group create \
    --name lobby \
    --platform paper \
    --platform-version 1.21.4 \
    --template lobby-base --template lobby-prod \
    --scaling-mode DYNAMIC \
    --min 1 --max 4 \
    --memory 2048 \
    --routing LOWEST_PLAYERS \
    --port-start 30000 --port-end 30100

Flags:

--name (required) — group identifier.
--platform (required) — paper, velocity, etc.
--platform-version — platform version string (e.g. 1.21.4).
--template <name> — template layers, applied in order. May be repeated.
--scaling-mode — STATIC, DYNAMIC, or MANUAL. Default DYNAMIC.
--min, --max — instance bounds. Defaults 1 and 10.
--memory — heap MB per instance. Default 1024.
--routing — routing strategy (e.g. LOWEST_PLAYERS).
--port-start, --port-end — port range to allocate from.

`group update`

prexorctl group update <name> --max 12 --memory 3072

Patches scaling, routing, or memory in place. Only flags you pass are sent; the rest stay as configured.

`group delete`

prexorctl group delete <name>

Stops every running instance in the group, then deletes it. Prompts for confirmation; pair with --json and a wrapping script if you need non-interactive deletes.

`group maintenance`

prexorctl group maintenance <name> on
prexorctl group maintenance <name> off

Toggles maintenance mode. While on, the group accepts no new players and the scheduler skips it.

`prexorctl instance`

Aliased to inst. Targets one or many running instances.

`instance list`

prexorctl instance list
prexorctl instance list --group lobby --state RUNNING

Flags:

--group <name> — filter by group.
--node <id> — filter by node.
--state <state> — filter by lifecycle state (SCHEDULED, PREPARING, STARTING, RUNNING, STOPPING, STOPPED, CRASHED, DRAINING).

`instance info`

prexorctl instance info <id>

Single-instance card with port, player count, memory, and uptime.

`instance start`

prexorctl instance start <group>

Asks the controller to schedule a new instance in the named group. The group’s scaling mode and current state determine how many instances are actually started; the response carries the count.

`instance stop`

prexorctl instance stop <id>
prexorctl instance stop <id> --force

Flags:

--force — SIGKILL immediately rather than the daemon’s graceful shutdown sequence.

`instance exec`

prexorctl instance exec <id> say Hello world
prexorctl instance exec <id> stop

Sends a command string to the server’s stdin via the daemon. Multi-word commands are joined with spaces.

`instance console`

prexorctl instance console <id>

Live SSE stream of the instance’s stdout/stderr, level-coloured. Press Ctrl-C to detach (the instance keeps running).

`prexorctl crash`

Crash reports are emitted automatically by the daemon when an instance exits unexpectedly.

`crash list`

prexorctl crash list
prexorctl crash list --group lobby --since 2026-05-01T00:00:00Z
prexorctl crash list --node node-1 --json

Flags:

--group <name> — filter by group.
--node <id> — filter by node.
--since <iso8601> — show crashes at or after the timestamp.

`crash info`

prexorctl crash info <id>

Shows context (instance, group, node, exit code, uptime, classification) and the tail of the instance’s log right before exit.

Scripting example

Restart every running instance in a group:

for id in $(prexorctl instance list --group lobby --state RUNNING --json \
              | jq -r '.[].id'); do
  prexorctl instance stop "$id"
done
prexorctl instance start lobby

Next up

Templates — managing the layers group create references with --template.
Group concepts — scaling modes, routing strategies, maintenance behaviour.
Instance lifecycle — what each state means.