pikasTech-unidesk/config/platform-infra/sub2api-codex-pool.yaml

pool:
  groupName: unidesk-codex-pool
  apiKeyName: unidesk-codex-pool-api-key
  apiKeySecretName: sub2api-codex-pool-api-key
  apiKeySecretKey: API_KEY
  minOwnerBalanceUsd: 1000
  defaultAccountPriority: 10
  defaultAccountCapacity: 10
  defaultAccountLoadFactor: 10
  defaultTempUnschedulable:
    enabled: false
    rules:
      - statusCode: 400
        keywords: [invalid_encrypted_content, encrypted content, could not be verified, could not be decrypted, bad_response_status_code, model_not_found, no available channel for model, unsupported, not supported, not support, 暂不支持, 可用模型]
        durationMinutes: 5
        description: Stable upstream 400 model-routing or Responses encrypted-content compatibility failures should use another account.
      - statusCode: 401
        keywords: [unauthorized, invalid api key, invalid_api_key, authentication, recovered upstream error]
        durationMinutes: 5
        description: Credential/auth failures should briefly cool down this account and use another account.
      - statusCode: 403
        keywords: [forbidden, access denied, quota, billing, capacity, weekly limit, less than 10% of your weekly limit left, run /status for a breakdown, recovered upstream error]
        durationMinutes: 5
        description: Permission, quota, or account-state failures should briefly cool down this account and use another account.
      - statusCode: 429
        keywords: [capacity, rate limit, rate_limit, quota, weekly limit, less than 10% of your weekly limit left, run /status for a breakdown, too many requests, overloaded, resource_exhausted, recovered upstream error]
        durationMinutes: 5
        description: Capacity and rate-limit responses are often temporary; briefly cool down this account and use another account.
      - statusCode: 500
        keywords: [capacity, overloaded, temporarily unavailable, temporary, upstream, recovered upstream error]
        durationMinutes: 5
        description: Transient upstream server failures should briefly cool down this account and prefer another account.
      - statusCode: 502
        keywords: [capacity, overloaded, temporarily unavailable, temporary, upstream, bad gateway, upstream request failed, unknown error, context deadline exceeded, context canceled, websocket dial, handshake response, recovered upstream error]
        durationMinutes: 5
        description: Gateway upstream failures, including recovered upstream error wrappers, should briefly cool down this account.
      - statusCode: 413
        keywords: [openai_error, payload too large, request too large, context length, context window, maximum context]
        durationMinutes: 5
        description: Large-context upstream failures should cool down the selected account so a larger-context channel can handle the request.
      - statusCode: 503
        keywords: [capacity, overloaded, temporarily unavailable, temporary, upstream, recovered upstream error, model_not_found, no available channel for model]
        durationMinutes: 5
        description: Service unavailable and upstream model-routing failures should briefly cool down this account.
      - statusCode: 504
        keywords: [gateway timeout, timeout, upstream, upstream request failed, unknown error, context deadline exceeded, context canceled, recovered upstream error]
        durationMinutes: 5
        description: Gateway timeout responses should cool down the selected account so another account can handle the next request.
      - statusCode: 524
        keywords: [timeout, a timeout occurred, cloudflare, gateway timeout, upstream, upstream request failed, unknown error, context deadline exceeded, context canceled, recovered upstream error]
        durationMinutes: 5
        description: Cloudflare 524 timeout responses should cool down the selected account so another account can handle the next request.
      - statusCode: 529
        keywords: [capacity, overloaded, temporarily unavailable, temporary, recovered upstream error]
        durationMinutes: 5
        description: Provider overloaded responses should briefly cool down this account and use another account.
profiles:
  entries:
    - profile: HY
      accountName: unidesk-codex-hy
      configFile: config.toml.HY
      authFile: auth.json.HY
      trustUpstream: true
      openaiResponsesWebSocketsV2Mode: off
      capacity: 10
      loadFactor: 10
      priority: 20
    - profile: thinkai
      accountName: unidesk-codex-thinkai
      configFile: config.toml.thinkai
      authFile: auth.json.thinkai
    - profile: dawclaudecode
      accountName: unidesk-codex-dawclaudecode
      configFile: config.toml.dawclaudecode
      authFile: auth.json.dawclaudecode
    - profile: ranmeng
      accountName: unidesk-codex-ranmeng
      configFile: config.toml.ranmeng
      authFile: auth.json.ranmeng
      capacity: 20
    - profile: gptclub
      accountName: unidesk-codex-gptclub
      configFile: config.toml.gptclub
      authFile: auth.json.gptclub
      trustUpstream: true
      capacity: 10
      priority: 100
    - profile: only
      accountName: unidesk-codex-only
      configFile: config.toml.only
      authFile: auth.json.only
      trustUpstream: true
      loadFactor: 1
      priority: 110
    - profile: zakuzaku
      accountName: unidesk-codex-zakuzaku
      configFile: config.toml.zakuzaku
      authFile: auth.json.zakuzaku
      priority: 10
    - profile: zakuzaku-x9nt
      accountName: unidesk-codex-zakuzaku-x9nt
      configFile: config.toml.zakuzaku-x9nt
      authFile: auth.json.zakuzaku-x9nt
      priority: 10
    - profile: ai2-hhhl
      accountName: unidesk-codex-ai2-hhhl
      configFile: config.toml.ai2-hhhl
      authFile: auth.json.ai2-hhhl
    - profile: freemodel
      accountName: unidesk-codex-freemodel
      configFile: config.toml.freemodel
      authFile: auth.json.freemodel
    - profile: nova-exellome
      accountName: unidesk-codex-nova-exellome
      configFile: config.toml.nova-exellome
      authFile: auth.json.nova-exellome
    - profile: socap
      accountName: unidesk-codex-socap
      configFile: config.toml.socap
      authFile: auth.json.socap
publicExposure:
  enabled: true
  proxyName: platform-infra-sub2api
  configMapName: sub2api-frpc-config
  deploymentName: sub2api-frpc
  frpcImage: fatedier/frpc:v0.68.1
  serverAddr: 74.48.78.17
  serverPort: 7000
  remotePort: 21880
  localIP: sub2api.platform-infra.svc.cluster.local
  localPort: 8080
  publicBaseUrl: https://sub2api.74-48-78-17.nip.io
  masterBaseUrl: http://127.0.0.1:21880
  masterFrps:
    configPath: /opt/hwlab-frp/frps.dev.toml
    containerName: hwlab-frps-dev
  masterCaddy:
    enabled: true
    domain: sub2api.74-48-78-17.nip.io
    configPath: /etc/caddy/Caddyfile
    serviceName: caddy
    upstreamBaseUrl: http://127.0.0.1:21880
    responseHeaderTimeoutSeconds: 600
    edgeRetry:
      enabled: true
      tryDurationSeconds: 10
      tryIntervalMilliseconds: 250
      retryMatch:
        methods: [POST]
        paths: [/responses/compact]
localCodex:
  backupSuffix: pre-sub2api
  providerName: OpenAI
  wireApi: responses
  supportsWebSockets: false
  responsesWebSocketsV2: false
  responsesSmokeModel: gpt-5.5
sentinel:
  monitor:
    enabled: true
  actions:
    enabled: true
  schedule: "*/1 * * * *"
  image: python:3.12-alpine
  sdk:
    openaiPythonVersion: "2.41.1"
  serviceAccountName: sub2api-account-sentinel
  configMapName: sub2api-account-sentinel-config
  credentialsSecretName: sub2api-account-sentinel-profiles
  stateConfigMapName: sub2api-account-sentinel-state
  cronJobName: sub2api-account-sentinel
  model: gpt-5.5
  endpoint: responses
  marker:
    prefix: UDSG_OK
    exact: true
  probe:
    timeoutSeconds: 30
    maxOutputTokens: 16
    transportRetryMinutes: 5
    userAgent: Go-http-client/1.1
  gatewayFailureMonitor:
    enabled: true
    lookbackSeconds: 900
    tailLines: 4000
    initialTtlMinutes: 5
    maxTtlMinutes: 30
    backoffMultiplier: 2
    paths:
      - /responses
      - /v1/responses
      - /responses/compact
      - /v1/responses/compact
  cadence:
    successInitialIntervalMinutes: 1
    successMaxIntervalMinutes: 20
    trustedSuccessMaxIntervalMinutes: 20
    untrustedSuccessMaxIntervalMinutes: 2
    successBackoffMultiplier: 2
    jitterPercent: 10
  freeze:
    initialTtlMinutes: 1
    maxTtlMinutes: 10
    backoffMultiplier: 2
    jitterPercent: 10
  pricing:
    usdPer1MInputTokens: 1.25
    usdPer1MOutputTokens: 10
  historyLimit: 200