pikopod up - pikopod

pikopod up [flags]

Starts two listeners: the agent on listen:agent_port (default 127.0.0.1:4700) and the sandbox on listen:sandbox_port (default 127.0.0.1:4600). Both share data_dir and fail independently. The spec watcher runs inside up for every upstream with a spec_source. Runs until SIGINT or SIGTERM, then drains for three seconds.

Flag	Meaning
`--wallclock-faults`	Every armed fault delays or hangs on the real wire. By default delays are virtualized and instant, which is what `scenario run` measures against.

Output

spec-declared enums: 1 field(s) across 1 upstream(s) keep their declared values readable on disk
spec watch ON: 1 declared source(s), re-checked every 1h0m0s
pikopod agent on http://127.0.0.1:4700  (upstreams: examplepay)
point your app's provider base URL at http://127.0.0.1:4700/<upstream>; /healthz shows progress
pikopod sandbox on http://127.0.0.1:4600  (sandboxes: examplepay)

Before serving, up lints every volatile_fields entry against what has been learned and prints any dead, stable or over-broad entry. See Volatile fields.

Refuses to start when

upstreams is empty.
listen is not loopback and no token is configured.
A sandbox has a webhook sink and its spec declares a signature, but the environment variable the envelope names is unset.
A volatile_fields entry is malformed.
The salt, the token file, or a config file holding a model key is readable by other users.

Every refusal names the fix. See Troubleshooting.

Observe, Running the sandbox, Deployment, Configuration.