Error Codes Reference
Complete error code reference: HTTP codes, messages, root causes, and solutions.
400 - Bad Request
Your request is malformed or has invalid parameters.
| Error | Cause | Solution |
|---|---|---|
INVALID_QUERY | Query too short (<3 chars) or long (>10K) | Reformulate: “How to validate tokens?” (5-100 chars recommended) |
INVALID_NAMESPACE | Namespace does not exist | Create with vectora namespace create your-ns |
INVALID_PARAMS | top_k is not a number, invalid filter | Check types: {"top_k": 10, "namespace": "proj"} |
MISSING_REQUIRED_FIELD | Missing required fields | Query and namespace are required |
Debug: Use vectora search --debug to see the exact payload.
401 - Unauthorized
Authentication or permission failure.
| Error | Cause | Solution |
|---|---|---|
INVALID_TOKEN | Expired or invalid token | Generate new: vectora auth token create |
MISSING_CREDENTIALS | API key not configured | vectora config set GEMINI_API_KEY "your-key" |
PERMISSION_DENIED | Your role does not have permission | Contact admin for upgrade (Owner/Admin can change RBAC) |
SESSION_EXPIRED | Session > 7 days | Reauthenticate: vectora auth login |
Debug: vectora auth status shows current token and expiration.
404 - Not Found
Resource does not exist.
| Error | Cause | Solution |
|---|---|---|
NAMESPACE_NOT_FOUND | Namespace deleted or not shared | List with vectora namespace list |
FILE_NOT_FOUND | File is not indexed | Reindex: vectora index --force |
CHUNK_NOT_FOUND | Chunk was deleted | No solution, chunk was removed |
429 - Rate Limited
You exceeded rate limit or API quotas.
| Error | Limit | Solution |
|---|---|---|
RATE_LIMIT_EXCEEDED | 60 req/min | Wait 1 min or implement exponential backoff |
QUOTA_EXCEEDED | Monthly token limit reached | Upgrade to Pro (500k tokens/month) or wait for reset |
CONCURRENT_LIMIT | Too many concurrent requests | Use connection pooling, max 10 concurrent |
Debug: Response headers show X-RateLimit-Remaining.
500 - Server Error
Internal error in Vectora or dependencies.
| Error | Cause | Solution |
|---|---|---|
EMBEDDING_FAILED | Voyage API unavailable | Check VOYAGE_API_KEY and Voyage status |
SEARCH_FAILED | Database unavailable | Check network connection, cluster status |
RERANK_FAILED | Reranker offline | Wait for automatic recovery (3s retry) |
LLM_ERROR | Gemini API error | Check GEMINI_API_KEY and quota on Google AI Console |
TIMEOUT | Request > 30s | Query too complex, simplify |
Debug: Full stack trace in ~/.vectora/logs/error.log.
CLI Exit Codes
0 Success
1 Generic error
2 Configuration error (invalid YAML file)
3 API error (500, timeout)
4 Auth failed (401, invalid token)
5 Namespace not found (404)
127 Command not foundEnable Debug Mode
For complex issues, debug mode provides a detailed view of the system’s inner workings, including network logs and raw API payloads.
# Debug mode shows stack traces and verbose logging
VECTORA_DEBUG=true vectora search "query"
# Or with flag
vectora --debug search "query"
# Or globally
vectora config set DEBUG trueExample: Common Troubleshoot
To facilitate resolution, keep your environment updated and check network settings before performing manual token exchange procedures.
Scenario: INVALID_TOKEN when using MCP in Claude Code
# 1. Verify if token exists
vectora auth status
# Output: Token: exp_...abc | Expires: 2026-05-19
# 2. If expired, regenerate
vectora auth token create
# Output: New token: exp_...xyz | Valid until: 2026-06-19
# 3. Update Claude Code config
# ~/.claude/claude_desktop_config.json
{
"mcpServers": {
"vectora": {
"command": "vectora",
"args": ["mcp", "--stdio"],
"env": {
"VECTORA_TOKEN": "exp_...xyz" # New token here
}
}
}
}
# 4. Restart Claude CodeReporting Bugs
If you see an undocumented error:
- Collect logs:
vectora logs --since 1h - Enable debug:
VECTORA_DEBUG=true vectora ... - Open issue: https://github.com/Kaffyn/Vectora/issues
- Include: error, stack trace, command used
Found an undocumented error here? Report it
Part of the Vectora ecosystem · Open Source (MIT)
External Linking
| Concept | Resource | Link |
|---|---|---|
| MCP | Model Context Protocol Specification | modelcontextprotocol.io/specification |
| MCP Go SDK | Go SDK for MCP (mark3labs) | github.com/mark3labs/mcp-go |
| Anthropic Claude | Claude Documentation | docs.anthropic.com/ |
| Voyage Embeddings | Voyage Embeddings Documentation | docs.voyageai.com/docs/embeddings |
| Voyage Reranker | Voyage Reranker API | docs.voyageai.com/docs/reranker |
| RBAC | NIST Role-Based Access Control Standard | csrc.nist.gov/projects/rbac |
Part of the Vectora ecosystem · Open Source (MIT) · Contributors