Start, stop and restart an instance

The power buttons in an instance's header start it, restart it and shut it down, as the power button on a physical machine would. Use them when the instance does not answer, after you change its settings, or to stop it while you do not need it.

An instance you shut down from inside its operating system shows as stopped within a minute. A stopped instance keeps its disk and addresses, and you pay for them until you cancel it. On an instance billed by use, stopping it ends the memory and CPU charges and keeps the disk charge (How Cloud Compute is billed). An instance with a plan price is billed in full while it is stopped.

• The instance is not suspended. A suspended instance shows This instance is suspended and no power buttons; select Open Billing to see why and settle it.

In the dashboard, go to Cloud Compute and open the instance. The Overview tab opens, and its Status card also has Start and Stop. The header shows the buttons that fit its state:

Start

Shown when the instance is stopped. Boots it.

Restart

Shown when it is running. Asks the operating system to shut down, then starts the instance again.

Stop

Shown when it is running. Asks the operating system to shut down cleanly. If it has not stopped after 60 seconds, we power it off.

Kill…

In the Instance actions menu beside the buttons, when the instance is running. Cuts the power at once.

A message confirms each action: Starting., Rebooting., Shutting down. or Powering off.. The status in the header changes a few seconds later.

To kill an instance, open the Instance actions menu, select Kill…, then select Kill instance in the dialog.

The Cloud Compute list has the same actions on each instance's card. A stopped instance has Start. A running instance has Restart, and the Power group of its ... menu holds Stop and Kill…. Start acts at once. Restart, Stop and Kill… ask first, with the buttons Restart instance, Stop instance and Kill instance. After you press one, the badge reads Starting…, Restarting… or Stopping… until the list reads a new state, and a second press sends nothing. A suspended, failed, setting up or moving instance has none of them (Find your instances).

A stopped instance billed by use pays for its disk only, so it does not hold its memory on its host. Start asks the host for that memory again. When the host has no room, the start is refused and the instance stays stopped with its disk safe. An instance that keeps a plan price has its place held, and Start is not refused for room.

A restart from the header stops the virtual machine and starts it again, and the new start picks up changes to its configuration: SSH keys, hostname, network settings and the display. A reboot typed inside the instance restarts only the operating system, so those changes wait for the next start from the dashboard.

The header shows the new status: running with the time it has been up, or stopped. The Cloud Compute list shows the same status on the instance's card, a moment after the action.

Stop or Restart does nothing

The operating system ignored the request to shut down. Stop powers the instance off after 60 seconds anyway. If a restart hangs, select Kill…, then Start.

Cannot perform power actions on a suspended instance

The service behind the instance is suspended. Settle it from the Billing tab; What happens when a payment fails explains suspension.

Instance is migrating

We are moving the instance to another host. Wait until the move finishes and try again.

There is no room to start this instance right now. Its disk is safe. Try again in a few minutes, or contact support if it stays like this.

The host has no room for the instance's memory. The answer is 409 with the code no_room_to_start. Try again in a few minutes.

Too many requests for this action. Try again later.

You sent more than 24 power actions for this instance in 60 seconds. Wait for the time the response gives in Retry-After.

• Use an instance's console to watch the instance boot.
• Repair an instance in rescue mode when it does not boot.
• Cloud Compute limits lists every rate limit.

POST /api/v1/client/vps/{uuid}/power sends a power action. Send action as one of:

start

Boot the instance.

reboot

Shut the operating system down and start it again. The dashboard's Restart.

shutdown

Shut down cleanly, and power off after 60 seconds if it has not stopped. The dashboard's Stop.

stop

Power off at once. The dashboard's Kill….

curl -X POST https://api.coritan.com/api/v1/client/vps/$INSTANCE_UUID/power \
  -H "Authorization: Bearer $CORITAN_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"action": "shutdown"}'

The request returns once the host has carried out the action, with {"status": "ok", "action": "shutdown"}. Any other word answers 400 with Invalid action. Must be: start, stop, shutdown, reboot. A suspended instance answers 403, one we are moving answers 409, a start on a host with no room answers 409 with the code no_room_to_start and the sentence in detail.message, and more than 24 actions in 60 seconds answer 429, as Rate limits describes.

GET /api/v1/client/vps/{uuid}/status reads the instance's live state from its host.

curl https://api.coritan.com/api/v1/client/vps/$INSTANCE_UUID/status \
  -H "Authorization: Bearer $CORITAN_TOKEN"

status

running or stopped as the host reports it, or migrating while we move the instance.

power_state

The machine's own state, such as running, paused or stopped.

cpu

CPU use as a fraction of the instance's vCPUs, from 0 to 1.

mem, maxmem

Memory in use and memory size, in bytes.

disk, maxdisk

Disk in use and disk size, in bytes.

netin, netout

Bytes received and sent since the instance last started.

uptime

Seconds since the instance last started.

agent_online

true when the QEMU guest agent inside the instance answered.

instance_status, rescue_mode, migration

The instance's status on your account, whether it is booted into rescue media, and the move in progress, if any.