Fixes #1047. `_doSingle`'s normal-relay path (cache disabled or cache miss on non-cachable request) ran `UrlFetchApp.fetch` → `getContent` → `base64Encode` with no error wrapper. Any throw — most commonly when the response body approaches Apps Script's ~50 MB ceiling and `base64Encode` blows the V8 heap, also URL-too-long / payload-too-large / quota exhaustion / 6-minute execution timeout — propagated unhandled, and Apps Script served its default `<title>Web App</title>` HTML error page in place of the JSON envelope. The Rust client (`parse_relay_json` in `domain_fronter.rs`) then failed to find JSON and surfaced the cryptic `bad response: no json in: <!DOCTYPE html>...` with no signal as to the actual cause. The reporter's symptom — a single failing host (`shc-dist.lostsig.co`, sonichacking.org) serving large ROM-hack binaries — matches this exactly. Every other download worked because they were all under the body-size ceiling. ## Fix Wrap the normal-relay block in `_doSingle` with `try { ... } catch (err) { return _json({ e: "fetch failed: " + String(err) }); }`. Mirrors the per-item try/catch already present in `_doBatch`. Turns the silent HTML crash into a structured `FronterError::Relay("fetch failed: …")` on the client side that pinpoints the real underlying error. Cache path intentionally untouched: - `_fetchAndCache` already wraps its own fetch in try/catch and returns `null` on any failure (so `_doSingle` falls through cleanly to the normal relay). - The cached-read path is bounded to ≤ `CACHE_MAX_BODY_BYTES` (35 KB) so it cannot trip the size limits that caused this bug. ## Verified locally on top of v1.9.22 - `node --check assets/apps_script/Code.gs`: clean ✅ - `cargo test --lib --release`: 209/209 ✅ (sanity — no Rust change) Reviewed via Anthropic Claude. Co-Authored-By: dazzling-no-more <noreply@github.com> Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Apps Script source
Three deploy-ready Apps Script files live here. They all speak the same {k, m, u, h, b, ct, r} wire protocol with mhrv-rs, so the client just points its script_id at whichever deployment you want — no mode change required.
Variants and origins
-
Code.gs— standard relay. Verbatim mirror of upstream. Apps Script does the outbound fetch itself. This is the default choice for most users.- Upstream: https://github.com/masterking32/MasterHttpRelayVPN/blob/python_testing/apps_script/Code.gs
- Credit: @masterking32. We do not modify this file.
- The mirror lives here so that (a) users on networks where
raw.githubusercontent.comis unreachable can still deploy from agit clone/ ZIP, and (b) we have a snapshot to diff against if upstream changes silently break the informal relay protocol.
-
CodeFull.gs— superset ofCode.gsthat additionally proxies raw-TCP / UDP viatunnel-node(used bymode: "full"). Maintained in this repo — written for this Rust port and not present upstream. Deploy this if you want full-tunnel mode; details in the file's header comment. -
Code.cfw.gs— Apps Script becomes a thin auth+forward layer; the actual outbound fetch happens on a Cloudflare Worker you also deploy (assets/cloudflare/). Derivative work — not unmodified upstream. The pattern of forwarding through a Cloudflare Worker came from denuitt1/mhr-cfw; this file inherits hardening fromCode.gs(decoy-on-bad-auth, fail-closed sentinels) and adds chunked batch forwarding (Promise.allon the Worker side,ceil(N/40)GAS calls per batch) that the upstreammhr-cfwdoes not have. Faster per-call latency, worse YouTube long-form, no fix for Cloudflare anti-bot. Readassets/cloudflare/README.mdbefore choosing this one.
What you must edit before deploying
For every variant: change AUTH_KEY from its placeholder to a strong secret, and use that same string in your mhrv-rs config's auth_key. Code.cfw.gs additionally requires setting WORKER_URL to your deployed Cloudflare Worker URL; CodeFull.gs additionally requires TUNNEL_SERVER_URL and TUNNEL_AUTH_KEY for the tunnel-node leg.