Local Linux

Chronicle’s released live-capture deployment is a local Linux host. Release binaries target x86_64-unknown-linux-gnu and aarch64-unknown-linux-gnu; the 0.1 release-verified runtime environment is Ubuntu 24.04/Linux 6.8/aarch64. x86_64 and other Linux 6.1+ environments require matching privileged acceptance; an artifact build is not runtime proof.

Readiness checklist

chronicle doctor

Live capture needs:

• release binaries target x86_64-unknown-linux-gnu and aarch64-unknown-linux-gnu;
• release-verified runtime: Ubuntu 24.04/Linux 6.8/aarch64;
• other Linux 6.1+ kernels and x86_64 require matching privileged acceptance;
• cgroup v2, BTF at /sys/kernel/btf/vmlinux, and required capabilities;
• CAP_BPF and CAP_NET_ADMIN for the recording process;
• embedded eBPF programs in the binary;
• a workload emitting bounded plaintext HTTP/1.1.

Use doctor --format json through the global option when an automation layer needs stable probe data:

chronicle --format json doctor

Local data and capacity

Recordings, WAL segments, manifests, checkpoints, and canonical payloads remain in the resolved local data directory. Omitted --duration means no whole-recording deadline; explicit deadlines are independent from bounded epoch/segment WAL limits. A parent recording may own many epochs without a total-WAL cap. Plan disk and file permissions around captured data, not just binary size.

Deployment boundaries

Chronicle does not currently ship Docker or Kubernetes packaging, an always-on distributed capture plane, PostgreSQL/S3 persistence, or remote artifact publication. Those are future concerns, not deployment instructions for 0.1.x.

For a supervised command, use:

chronicle record --name checkout -- ./my-app

For an existing process or cgroup, use --pid PID or --cgroup PATH; Chronicle does not terminate those workloads. Replay remains loopback-only and explicitly authorized.