Why I Chose Detached ECDSA P-256 Signatures for Production OTA

I used to think a firmware release was the binary produced after a successful build. Production needed a release artifact that the server and device could verify without embedding a private signing secret into either runtime.

The LOUP firmware path had a particularly unforgiving constraint: SIP/audio behavior already worked well enough to be valuable, so release work could not casually erase NVS, rewrite the flash layout, merge unrelated features or replace the rollback image. The goal was to make delivery safer without sacrificing the only known-good behavior.

For this article, the key evidence is specific: The current OTA policy specifies detached ECDSA P-256/SHA-256 signatures for production application artifacts and identifies the verification key separately from the private signing key. The retained result was equally specific: Only clean-source, cryptographically verified application releases may enter OTA-servable states. I treat both as observations from the documented release state, not as universal ESP32 rules.

Case notebook

The table is intentionally stricter than a normal release note. A release note usually tells a reader what changed. This notebook also records what was not proven and which shortcut would have produced a misleading green status. That distinction mattered repeatedly in the LOUP work, especially while the golden V132A artifact, reconstructed source tree, V133A feature branch and later OTA control plane existed at different maturity levels.

The OTA trust chain has several independent links

HTTPS authenticates the transport endpoint and protects the connection. A SHA-256 digest detects content mismatch. An ECDSA signature proves that an authorized signing key approved the artifact. Compatibility metadata prevents a correctly signed artifact from being sent to the wrong hardware or persistent-state generation. Bootloader security can later enforce additional device-local policy.

Those controls should stay separate in both implementation and incident response. If a checksum mismatches, I investigate corruption or artifact substitution. If the signature fails, I investigate release authorization or key mismatch. If compatibility fails, the artifact may be perfectly authentic but still unsafe for that device. If TLS fails, I do not bypass verification just because the signature exists.

Key custody is part of the same architecture. The private key authorizes executable code, so the runtime OTA server does not need it merely to serve and verify releases. Keeping signing authority off the distribution host limits the consequence of one class of server compromise.

A concrete scenario I use to test the rule

Detached signatures are convenient because the application binary remains an ordinary binary. The release system can carry signature/key metadata beside it without inventing a custom executable container.

For this article, the scenario is useful because it targets the rejected shortcut directly: Treating HTTPS transport as sufficient proof that an uploaded firmware binary was produced by the release authority.. I want the system to make that shortcut either impossible or obviously non-compliant with the release gate.

The falsification question is equally important. If a future implementation can demonstrate the same safety property with a simpler mechanism, I would change the mechanism. What I would not change casually is the invariant: Transport security protects the connection; artifact signing protects the software supply chain.. The release process exists to preserve that invariant while the implementation evolves.

The mechanism underneath the release decision

OTA is remote code execution by design, so release authority is part of the threat model. TLS, checksums, signatures, compatibility metadata and irreversible eFuse policy solve different problems. I avoid using one of those controls as a substitute for the others.

For Why I Chose Detached ECDSA P-256 Signatures for Production OTA, the important mechanism is this: Detached signatures allow the binary to remain the normal ESP32 application artifact while signature metadata travels alongside it and can be verified before installation.

That mechanism tells me which evidence is relevant. If the question is artifact identity, a call test alone is not enough; I need a hash and source identity. If the question is rollback, a signature alone is not enough; I need partition and persistent-schema compatibility. If the question is promotion, a CI pass is not enough; I need observed device state tied to the same release ID.

operator machine:
  artifact -> SHA-256 -> ECDSA P-256 signature
                     private key stays here

OTA server:
  stores artifact + detached signature + public verification metadata

device:
  verifies assigned artifact before accepting it

Integrity, authorization and transport trust remain distinct layers.

I use this model to stop release engineering from becoming a sequence of shell commands. The commands are implementation. The release contract is the set of invariants that must still be true when the commands finish.

The controls I would require before accepting this state

• verify TLS without insecure fallback
• verify SHA-256 content identity
• verify detached release signature
• keep private signing material off the runtime host
• validate model/hardware/partition/schema compatibility

The point is not to maximize checklist length. Each control closes a different ambiguity that appeared in the real work. For this case, the shortcut I reject is Treating HTTPS transport as sufficient proof that an uploaded firmware binary was produced by the release authority.. If that shortcut is allowed, the release can look successful while the underlying recovery or provenance guarantee is false.

I prefer a release gate that fails loudly and leaves the old artifact usable. That is why full-flash erasure, force-pushing golden tags, disabling certificate verification, or widening a rollout to compensate for unclear state are all wrong directions. They destroy evidence or increase blast radius exactly when uncertainty is highest.

How I would try to break this before trusting it

Release safety is difficult to prove with only the happy path. For this class of change I want at least one test that intentionally violates the assumption the release depends on.

For Why I Chose Detached ECDSA P-256 Signatures for Production OTA, I would construct a negative test around the mechanism: Detached signatures allow the binary to remain the normal ESP32 application artifact while signature metadata travels alongside it and can be verified before installation. That might mean removing a required source file from the clean checkout, presenting a wrong signature, assigning a release to the wrong hardware revision, forcing first-boot self-test failure, rolling back after a schema migration, or replaying a terminal OTA result for an older release ID.

The expected behavior should be boring: reject the artifact or assignment, keep/restore the previous accepted image, preserve persistent state where promised, and surface an attributable failure state. A test is especially valuable when it proves the system does not accept a dangerous shortcut.

This is why I distinguish recoverability tests from build tests. A compiler can prove syntax and linking. It cannot prove that a power interruption during slot write, a bad first boot or a stale heartbeat result leaves the fleet in a state the operator can understand.

Keep chronology honest

The V132A/V133A recovery documents are useful because they did not retroactively turn incomplete work into completed work. At the August handoff, the golden V132A rollback binary was proven, while the V133A feature branch had source work that still needed clean-build, flash, physical power and clear-audio regression evidence. Later fleet/OTA work added a different layer of production acceptance.

That chronology matters for this topic because Only clean-source, cryptographically verified application releases may enter OTA-servable states. should be read as the result of the evidence chain that actually existed at that stage. I do not use a later production capability to rewrite an earlier handoff as if it had already passed.

This is also how I want release dashboards and articles to behave. A state such as built, signed, assigned, booted, accepted, active or stable should mean one thing and be backed by the evidence required for that state. The system becomes hard to operate when success words float free of their acceptance gates.

What changes when the fleet grows

On one board, I can recover with USB and inspect serial output manually. At five or twenty devices, release identity, assignment and rollback have to be queryable. At a larger production fleet, I would tighten the same model rather than replace it: hardware-backed key provisioning, Secure Boot v2 and Flash Encryption under a controlled ceremony, more formal signing custody, staged rollout cohorts, automated rollback evidence, schema-compatibility tests and independent release audit records.

The important thing is that scale does not remove the original invariant. Transport security protects the connection; artifact signing protects the software supply chain. A larger fleet only makes violations more expensive.

I would also keep service-mode releases separate from normal app OTA. Partition-table, bootloader and security-epoch transitions deserve a dedicated recovery plan because they modify the machinery that normal OTA depends on. That separation is useful at ten devices and essential at ten thousand.

The lesson I keep

The result I keep from this case is: Only clean-source, cryptographically verified application releases may enter OTA-servable states.

The deeper lesson is Transport security protects the connection; artifact signing protects the software supply chain.

That is how I now define production firmware work. The release is not the moment a .bin file appears. It is the chain that connects source, artifact, signature, compatibility, flash topology, persistent state, real-device acceptance and observable running state—with a rollback path whose assumptions have actually been tested.