iflytek/astron-agent · error · BusinessException

MODEL_CHECK_FAILED

MODEL_CHECK_FAILED

Error message

MODEL_CHECK_FAILED

What it means

Thrown in BotChatServiceImpl.debugChatMessageBot when modelService.checkModelBase rejects the resolved model's llmId/serviceId/url for the requesting uid/spaceId. It wraps ResponseEnum.MODEL_CHECK_FAILED, meaning the model failed a base availability/authorization check before the chat task was built.

Solutions

  1. Confirm the bot's configured model still exists and is deployed (check llmId/serviceId/url).
  2. Verify the user (uid) has access to the model in spaceId; grant model permissions if needed.
  3. Re-select a valid model in the bot configuration and save.
  4. Check model service health/logs for why checkModelBase returned false.

Example fix

// before
if (!modelService.checkModelBase(modelConfig.llmInfoVo().getLlmId(),
        modelConfig.llmInfoVo().getServiceId(), modelConfig.llmInfoVo().getUrl(), request.getUid(), request.getSpaceId())) {
    throw new BusinessException(ResponseEnum.MODEL_CHECK_FAILED);
}
// after
if (!modelService.checkModelBase(modelConfig.llmInfoVo().getLlmId(),
        modelConfig.llmInfoVo().getServiceId(), modelConfig.llmInfoVo().getUrl(), request.getUid(), request.getSpaceId())) {
    log.error("Model check failed, llmId: {}, uid: {}, spaceId: {}",
            modelConfig.llmInfoVo().getLlmId(), request.getUid(), request.getSpaceId());
    throw new BusinessException(ResponseEnum.MODEL_CHECK_FAILED);
}
Defensive patterns

Strategy: validation

Validate before calling

// Pre-flight check before debug chat:
boolean ok = modelService.checkModelBase(llmId, serviceId, url, uid, spaceId);
if (!ok) { throw new BusinessException(ResponseEnum.MODEL_CHECK_FAILED); }

Try / catch

try {
    botChatService.debugChatMessageBot(request, sseEmitter);
} catch (BusinessException e) {
    if (ResponseEnum.MODEL_CHECK_FAILED == e.getCode()) {
        // prompt user to re-select an available model for this space
    }
}

Prevention

When it happens

Trigger: debugChatMessageBot resolves a modelConfig; if modelConfig != null and checkModelBase(llmId, serviceId, url, uid, spaceId) returns false (model unavailable, not deployed, or not permitted in that space), BusinessException(MODEL_CHECK_FAILED) is thrown.

Common situations: Bot configured with a model the user/space has no access to; model service URL changed or deployment is down; model was unpublished/removed but still referenced in bot config; cross-space usage of a private model.

Understand the failure class

Background: "Invalid value" and "allowed values are" config errors: what your library rejected and how to fix it — this error's family across 41 libraries.

Related errors

AI-assisted analysis of iflytek/astron-agent@5e758547a8 (2026-09-12). Data as JSON: /api/errors/b55ae8be73c72e1c. Report an issue: GitHub.

Appendix: source

Thrown at console/backend/hub/src/main/java/com/iflytek/astron/console/hub/service/chat/impl/BotChatServiceImpl.java:253

     * @param request Debug chat bot request parameters
     * @param sseEmitter SSE emitter
     * @param sseId SSE ID
     */
    @Override
    public void debugChatMessageBot(DebugChatBotReqDto request, SseEmitter sseEmitter, String sseId) {
        try {
            List<SparkChatRequest.MessageDto> messageList;
            // get personality config prompt
            String prompt = personalityConfigService.getChatPrompt(request.getPersonalityConfig(), request.getPrompt());
            ModelConfigResult modelConfig = resolveChatModelConfiguration(
                    request.getModelId(), request.getModel(), request.getUid(), request.getSpaceId(), sseEmitter);
            int maxInputTokens = modelConfig == null ? this.maxInputTokens : modelConfig.maxInputTokens();
            messageList = buildDebugMessageList(request.getText(), prompt, request.getMessages(), maxInputTokens,
                    request.getMaasDatasetList());
            if (modelConfig != null) {
                if (!modelService.checkModelBase(modelConfig.llmInfoVo().getLlmId(),
                        modelConfig.llmInfoVo().getServiceId(), modelConfig.llmInfoVo().getUrl(), request.getUid(), request.getSpaceId())) {
                    throw new BusinessException(ResponseEnum.MODEL_CHECK_FAILED);
                }
            }
            AgentChatTask task = AgentChatTask.builder()
                    .llmInfoVo(modelConfig == null ? null : modelConfig.llmInfoVo())
                    .sparkModelName(modelConfig == null ? request.getModel() : null)
                    .messages(messageList)
                    .openedTool(request.getOpenedTool())
                    .mcpServerUrls(request.getMcpServerUrls())
                    .skills(enrichBotSkills(request.getSkills()))
                    .tools(request.getTools())
                    .workflows(request.getWorkflows())
                    .userId(request.getUid())
                    .chatId(null)
                    .botId(request.getBotId())
                    .spaceId(request.getSpaceId())
                    .debugSessionId(request.getDebugSessionId())
                    .rawUserText(request.getText())
                    .chatReqRecords(null)

View on GitHub (pinned to 5e758547a8)