pikasTech-unidesk/docs/reference/master-server-ops.md

Master Server Ops

This document records master-server architecture and decision rules. Installed Codex wrappers own their command syntax; this reference owns profile boundaries, safety rules and acceptance criteria. General server lifecycle and GC commands remain in the unidesk-ops skill. Do not place secrets, one-off incident logs, or dated changelog notes here.

Execution Boundary

• The master server is the production entry point and control plane. Default work on this host is limited to light source edits, Git operations, status checks, health checks, logs, diagnostics, JSON CLI control actions, and local profile wrapper maintenance.
• Heavy builds and repository-wide checks should run on an external execution surface such as G14, D601, CI/CD, or the target runtime.
• The local UniDesk checkout at /root/unidesk can be used for lightweight UniDesk CLI/trans/tran/helper changes.

Codex Provider Profiles

mycx and oncx are stable system Codex wrappers under /usr/local/bin/; both use the default /root/.codex home and are described in the shared-state subsection below. dscx, mxcx, and acx are isolated local Codex profile wrappers under /root/.local/bin/. They are host-level tools, not UniDesk service code. acx is the unified multi-model entry; gocx remains the OpenCode Zen Go compatibility entry, and older single-model dscx-go, dfcx-go, and glcx-go wrappers are compatibility entries.

• dscx uses CODEX_HOME=/root/.codex-deepseek-v4-pro, model deepseek-v4-pro, Codex custom provider deepseek, and local Moon Bridge at http://127.0.0.1:38440/v1.
• mxcx uses CODEX_HOME=/root/.codex-minimax-m3, model MiniMax-M3, Codex custom provider minimax, and local Moon Bridge at http://127.0.0.1:38441/v1.
• acx uses CODEX_HOME=/root/.codex-acx and default model gpt-5.5-only. GPT aliases use Codex custom providers that connect directly to their OpenAI-compatible Responses upstreams; OpenCode Zen Go aliases still use the local ACX router only to reach gocx/Moon Bridge.
• gocx uses CODEX_HOME=/root/.codex-opencode-go-all, default model glm-5.1, Codex custom provider opencode, and local Moon Bridge at http://127.0.0.1:38447/v1.
• dscx-go uses CODEX_HOME=/root/.codex-opencode-go, model deepseek-v4-pro, Codex custom provider opencode, and local Moon Bridge at http://127.0.0.1:38443/v1.
• dfcx-go uses CODEX_HOME=/root/.codex-opencode-flash, model deepseek-v4-flash, Codex custom provider opencode, and local Moon Bridge at http://127.0.0.1:38444/v1.
• glcx-go uses CODEX_HOME=/root/.codex-opencode-glm, model glm-5.1, Codex custom provider opencode, and local Moon Bridge at http://127.0.0.1:38446/v1.
• acx includes all OpenCode Zen Go upstream slugs plus gpt-5.5-only and gpt-5.5-sub2api in one model-catalog.json so Codex can use /model or -m <model> within the same profile.
• acx routes OpenCode Zen Go models to the existing gocx Moon Bridge. GPT models must not pass through Moon Bridge or the ACX router: gpt-5.5-only uses the direct only OpenAI-compatible Responses endpoint and gpt-5.5-sub2api uses the Sub2API pool endpoint, both with upstream model gpt-5.5.
• GPT direct providers must receive their API key through an environment-key path such as ACX_GPT_DIRECT_API_KEY, read from the matching /root/.codex/auth.json.<profile> file by the wrapper. Do not let direct GPT calls fall back to /root/.codex-acx/auth.json; that file may contain only the local-router dummy key.
• Isolated Moon Bridge and router wrappers read upstream API keys from profile auth files or, where the compatibility path still requires it, wrapper-injected environment variables; generated runtime configs live under the profile .tmp/ directory with mode 0600. This compatibility rule does not apply to mycx Pika auth, whose command-backed contract is defined below. Do not copy upstream keys into documentation.
• Each profile must include model_catalog_json in config.toml pointing to a profile-local model-catalog.json entry for its active model. Missing catalog metadata causes Codex to fall back to default metadata, which lowers the effective context window and prints Model metadata ... not found.
• Profile context metadata must match the intended upstream limit closely enough for Codex auto-compact to fire before provider rejection. Keep the local profile metadata in sync with the actual model family you are routing to.
• Current master-server profile baselines:
  • GPT profiles exposed through acx use model_context_window = 272000 and model_auto_compact_token_limit = 240000. This represents the Codex-facing input window for GPT-5.5, not the larger raw API model window.
  • deepseek-v4-pro and deepseek-v4-flash use model_context_window = 1000000 and model_auto_compact_token_limit = 900000.
  • Other local Moon Bridge profiles, including glm-5.1, MiniMax-M3, and the non-DeepSeek OpenCode models exposed through acx/gocx, use model_context_window = 200000 and model_auto_compact_token_limit = 180000.
• Keep the wrapper-generated Moon Bridge/router metadata aligned with the profile config.toml and model-catalog.json. If these diverge, Codex and the local admission layer may disagree about compaction and request size behavior.
• hyueapi.com / .hyueapi.com must remain in NO_PROXY / no_proxy for Codex API channels.

Shared CODEX_HOME Wrappers (mycx / oncx)

mycx and oncx reuse the same system codex binary and the same /root/.codex state tree. Their stable entrypoints are native executable wrappers, not shell functions:

• /usr/local/bin/mycx selects profile pika; its writable profile layer is /root/.codex/pika.config.toml.
• /usr/local/bin/oncx selects profile only; its writable profile layer is /root/.codex/only.config.toml.
• /root/.codex/config.toml is the shared base layer, while /root/.codex/state_5.sqlite and /root/.codex/sessions/**/rollout-*.jsonl hold the shared session index and durable transcripts.
• The Webterm declaration in config/platform-infra/webterm.yaml starts mycx; wrapper path or startup changes must preserve that original entrypoint and include a Webterm launch check.
• codex-pool configure-local owns only the unsuffixed shared config.toml and auth.json consumer files described in docs/reference/platform-infra.md. It does not own the profile-v2 pika.config.toml, only.config.toml, or auth.json.pika files and must not regenerate them as Sub2API consumer state.
• Shell startup files must not define same-name functions that shadow these executables or regenerate profile files on launch. After changing a wrapper, existing Codex processes must be exited and existing shells must clear stale functions or start a new shell before validation.

For profile-v2, /model persists model and reasoning effort to the active profile file. The two wrappers therefore remember their last selections independently. -m and -c model... / -c model_reasoning_effort... are runtime overrides and must not be treated as persistent model selection. A fresh same-profile restart, a switch between profiles, and resuming an existing session are distinct behaviors and must be diagnosed separately.

Session discovery has an additional identity boundary:

• Sharing CODEX_HOME does not guarantee identical /resume results. The picker also filters by the effective provider ID and by cwd, source, and archive state.
• Provider IDs are persisted, case-sensitive session bucket keys. The active wrappers use distinct provider identities, and historical sessions can retain earlier capitalization or provider names.
• In the deployed picker behavior, --all removes the cwd restriction but does not remove the provider restriction. Revalidate this contract after Codex upgrades before changing wrappers or migrating history.
• Resume uses the current profile's effective model, provider, and reasoning overrides; it does not necessarily restore the model last recorded by the selected session. Fresh model persistence and session-model restoration must therefore have separate acceptance criteria.
• Rollout JSONL is durable history and can repopulate the SQLite index. Never change only state_5.sqlite to merge provider buckets. A reviewed history migration must stop active Codex processes, update rollout metadata with a structured JSON parser, update the SQLite index transactionally, preserve provenance, and verify the same session ID set after reindexing.

Provider identity is a compatibility contract. Any change to CODEX_HOME, profile name, provider ID, or provider-ID capitalization must preflight existing history and warn or block when it would create a new hidden bucket. Do not make different real backends impersonate one provider solely to merge picker lists. The preferred future unification is a picker policy with explicit current, all, or configured aliases, visible origin-provider metadata, and an explicit continue/fork decision for cross-provider resume.

Pika authentication must use command-backed provider auth from [model_providers.Pika.auth], with the credential source /root/.codex/auth.json.pika restricted to mode 0600. Do not export the Pika API key into the wrapper or Codex environment: inherited environment variables can be captured by shell snapshots and propagated to child tools. Shell snapshots must also be restricted to mode 0600 and must not contain reusable credentials. Treat any plaintext snapshot occurrence as credential exposure, remove the copy, and rotate the credential.

Diagnostics and validation must remain secret-safe:

• Report only effective CODEX_HOME, wrapper path, active profile file, provider ID, model/reasoning, session counts by provider, credential presence or fingerprint, and relevant file modes.
• Verify each profile by selecting a non-baseline /model, exiting, fresh-starting the same wrapper, and confirming the selection remains while the other profile file is unchanged.
• Verify a temporary CLI override does not rewrite the profile, and verify Pika command-backed auth works with the corresponding API-key environment variable absent.
• When /resume differs, compare CODEX_HOME, provider ID, cwd, source, and archive filters before attributing the result to storage loss or caching.

The investigation evidence and source-level rationale for these rules are retained in UniDesk issue #1640; this document is the authority for the active long-term contract.

Moon Bridge

Moon Bridge is installed as /root/.local/bin/moonbridge. The binary exposes OpenAI Responses endpoints and bridges Codex to provider-specific upstream APIs. The installed profile wrappers own bridge command syntax; this section records the supported runtime behavior.

The local source copy for the installed patched build is /root/src/moon-bridge-sanitize. When changing Moon Bridge, keep the change narrow, run package-level tests, build a replacement binary, back up the previous binary, then restart affected profile bridges. Default master-server build restrictions still apply.

Profile architecture:

• dscx bridge-start renders profile config and starts Moon Bridge on 127.0.0.1:38440.
• mxcx bridge-start renders profile config and starts Moon Bridge on 127.0.0.1:38441.
• acx route-start renders the ACX router config and starts the local routing service on 127.0.0.1:38448 for non-GPT ACX aliases that still need the OpenCode Zen Go bridge path.
• gocx bridge-start renders multi-model OpenCode Zen Go profile config and starts Moon Bridge on 127.0.0.1:38447.
• dscx-go bridge-start renders profile config and starts Moon Bridge on 127.0.0.1:38443.
• dfcx-go bridge-start renders profile config and starts Moon Bridge on 127.0.0.1:38444.
• glcx-go bridge-start renders profile config and starts Moon Bridge on 127.0.0.1:38446.
• The wrappers start Moon Bridge with setsid and a profile-local PID file.
• Logs are written under <CODEX_HOME>/logs/moonbridge/.
• dscx routes DeepSeek through Moon Bridge using Anthropic-compatible upstream + deepseek_v4 extension.
• mxcx routes MiniMax through Moon Bridge using openai-response upstream passthrough.
• acx routes OpenCode Zen Go models to gocx/Moon Bridge. gocx, dscx-go, dfcx-go, and glcx-go route OpenCode Zen Go through Moon Bridge using openai-chat upstream at https://opencode.ai/zen/go/v1/chat/completions. The Codex side remains wire_api = "responses" against the local Moon Bridge URL.
• acx routes GPT aliases directly to OpenAI-compatible Responses endpoints and must not send GPT traffic through Moon Bridge or through the local ACX router. In GPT mode, acx status should report mode=gpt-direct, routerRequired=false, and no listener on 127.0.0.1:38448.
• OpenCode Zen Go model IDs must use the upstream slug, such as glm-5.1; display names such as GLM-5.1 are not profile model identifiers.
• Do not keep local handwritten bridge scripts, static alternate moonbridge.config.yml files, or other sidecar proxy paths for OpenCode Zen Go profiles. The only supported runtime path is wrapper-generated .tmp/moonbridge.generated.yml plus /root/.local/bin/moonbridge.
• For OpenCode Zen Go profiles, set an explicit user_agent in the generated Moon Bridge provider config. The upstream may reject default client signatures.
• Moon Bridge now performs a local context-window admission check on the Responses path before forwarding oversized requests upstream. The expected failure shape for an oversized prompt is local HTTP 400 with invalid_request_error and code context_length_exceeded, not an upstream-translated 502 Bad Gateway.
• Do not switch MiniMax to openai-chat for Codex CLI unless tool-enabled smoke proves it works.

Profile validation:

• *-go raw-smoke verifies the upstream OpenCode Zen Go Chat Completions API directly.
• *-go bridge-smoke verifies local Moon Bridge's /v1/responses translation path.
• *-go exec '在吗' verifies the actual Codex profile. Passing output must not contain Model metadata ... not found; latest session records should show model_context_window derived from the profile catalog, not fallback metadata.
• acx status, acx models, acx gpt-only exec '在吗', acx gpt-sub2api exec '在吗', and default acx exec '在吗' verify GPT direct mode. Passing GPT verification should show a real Codex Responses turn and, for repeated or resume traffic, nonzero cached_input_tokens; it should not require a listener on 127.0.0.1:38448.
• For OpenCode Zen Go aliases exposed through acx, use acx route-start, acx route-status, acx models, and acx -m <opencode-model> exec '在吗' to verify the router-to-gocx path.
• gocx raw-smoke [model], gocx bridge-smoke [model], and gocx -m <model> exec '在吗' verify specific OpenCode Zen Go models. Omitting [model] uses the default glm-5.1.
• ReasoningSummaryDelta without active item in Codex stderr is a separate adapter noise from reasoning summary events. It is not the same failure as missing model metadata and does not by itself prove the profile is unusable.

OpenAI Responses Tool Argument Sanitizer

Moon Bridge owns the online sanitizer for MiniMax-style bad tool argument JSON on the openai-response passthrough path. This is the first line of defense; wrapper-level session cleanup is only for JSONL that was already corrupted before the sanitizer was installed.

Sanitizer rules: recursively scans ResponsesRequest.input, repairs tool-call arguments before upstream, repairs response-side arguments in completed events (not delta chunks), falls back to {} on unrecoverable input. Response rewriting drops upstream Content-Length.

MiniMax Session Recovery

mxcx includes a cleanup and guard layer for corrupted MiniMax-backed session JSONL. The wrapper owns recovery command syntax; this section documents behavioral rules.

• mxcx resume auto-runs session-clean + session-guard before invoking Codex. Uses CODEX_HOME=/root/.codex-minimax-m3 profile.
• session-clean is strictly scoped to invalid tool-call arguments: malformed JSON, MiniMax sentinel text, and schema-invalid cases. Must preserve line order, non-tool messages, reasoning, outputs, token records, and session metadata. Must not compact/summarize/truncate/reorder transcript.
• New invalid arguments should be prevented online by Moon Bridge sanitizer. Use session-clean only to recover already-written sessions.
• session-guard injects mxcx-minimax-tool-and-apply-patch-guard-v2, avoiding duplicates.

MiniMax Apply-Patch Operations

MiniMax-backed sessions must use the same UniDesk remote text patch contract as other agents: route first, operation second, and apply-patch v2 by default. The stable write shape is trans <provider>:/absolute/workspace apply-patch < patch.diff; read-only inspection may use trans <provider>:/absolute/workspace nl -ba file or equivalent bounded commands.

• If apply-patch reports failed to find expected lines, first read the exact current target block, then retry with a smaller Update File hunk, an @@ <unique anchor> hint, or multiple small hunks. This is normal stale-context recovery, not a reason to switch tools.
• Do not recover text patch failures by using download / upload, remote Python/Perl/sed heredocs, cat > / tee whole-file rewrites, or apply-patch-v1, unless apply-patch itself is unavailable or the target is non-text / bulk mechanical generated content.