Skip to main content
Use this endpoint to restore a paused capsule from its snapshot. Wrenn boots a fresh Firecracker VM, loads the saved filesystem, and uses userfaultfd (UFFD) for lazy memory page loading so the capsule is ready to accept commands quickly. The capsule resumes exactly where it left off — running processes, in-memory state, and all.

Endpoint

Authentication

Pass your API key in the X-API-Key header.

Path parameters

string
required
The ID of the capsule to resume (e.g., cl-abc123). The capsule must be in paused status.

Response

Returns 200 OK with the updated capsule object. The status field will be "running" once the VM has booted and the internal agent is ready.
string
required
Unique capsule ID, prefixed with cl-.
string
required
Will be "running" after the capsule has fully resumed.
string
required
Name of the template the capsule was created from.
integer
required
Number of virtual CPUs allocated.
integer
required
Memory allocated, in MiB.
integer
required
Inactivity TTL in seconds.
string
required
ISO 8601 timestamp of when the capsule was originally created.
string
required
ISO 8601 timestamp of when the capsule last changed status.
string
ISO 8601 timestamp of when the capsule most recently reached running status.

Example response

curl

Python SDK

The SDK handles resume transparently when you call Capsule.connect() on a paused capsule:
You can also resume explicitly on an existing instance:

Error responses