Skip to content

Search API (api-skillnet.openkg.cn) returns 502 Bad Gateway — backend appears down #28

Description

@Ruofei-personal

Describe the bug

The SkillNet Search API at http://api-skillnet.openkg.cn is returning 502 Bad Gateway for every request. The Python SDK (skillnet-ai, SkillNetSearcher.search()) consequently fails for any query, since it calls http://api-skillnet.openkg.cn/v1/search.

The nginx reverse proxy is responding, but the upstream application behind it appears to be down (crashed / not running), which is what produces the 502.

To Reproduce

curl -i "http://api-skillnet.openkg.cn/v1/search?q=ppt&mode=keyword&limit=20"

Or via the SDK:

from skillnet_ai.searcher import SkillNetSearcher
SkillNetSearcher().search(q="ppt", mode="keyword")   # raises HTTPError 502

Current behavior

HTTP/1.1 502 Bad Gateway
Server: nginx/1.10.3 (Ubuntu)

<html><head><title>502 Bad Gateway</title></head>
<body><center><h1>502 Bad Gateway</h1></center>
<hr><center>nginx/1.10.3 (Ubuntu)</center></body></html>

Diagnosis / scope

Check Result
GET /v1/search?q=... (keyword & vector) 502
Every other path (/, /health, /docs, /v1, /openapi.json) 502 (whole backend down, not a single route)
Response server header nginx/1.10.3 (Ubuntu) — proxy is up, upstream is not
DNS for api-skillnet.openkg.cn resolves fine -> 121.41.117.246
Front-end site http://skillnet.openkg.cn/ 200 OK (only the API backend is affected)
API host in current README still http://api-skillnet.openkg.cn/v1/search (so this is the intended endpoint)

This points to the API service process being down rather than a client/config issue — the nginx 502 means the proxy can't reach its upstream.

Expected behavior

GET /v1/search returns 200 with a JSON SearchResponse.

Environment

  • skillnet-ai version: 0.0.16
  • Date/time observed: 2026-06-29 (UTC+8)
  • Network: reproduced from a clean client, no proxy

Additional note

A couple of minor robustness suggestions for the service side, if useful:

  • A lightweight /health endpoint that stays up independent of the main app would make outages like this easier to detect.
  • Auto-restart (e.g. systemd/supervisor) on the API process would reduce downtime from crashes.

Thanks for maintaining SkillNet! Happy to provide more details or test once it's back up.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions