Skip to main content
Glama

DevDocs MCP Server

by cyberagiinc
fix_job_state_mixing.md2.04 kB
# BUG: URLs from Previous Job Appear in New Crawl **Issue:** After starting a new discovery/crawl (e.g., for `docs.crawl4ai.com`), URLs from a completely different, previously stopped job (e.g., `clerk.com`) are appearing in the Crawl Queue and potentially being processed. **Observed Behavior:** (See user-provided image) The Crawl Queue shows a mix of URLs from the new target (`docs.crawl4ai.com`, status 'pending crawl') and the old target (`clerk.com`, status 'completed'). **Expected Behavior:** Starting a new discovery should completely clear the state (Job ID, URL list, statuses) associated with any previous job, showing only the URLs relevant to the *new* discovery and crawl process. **Affected Components:** * `app/page.tsx` (State management, localStorage clearing, `handleSubmit` logic) * `backend/app/status_manager.py` (Potentially, if job states are not isolated) * LocalStorage persistence logic **Potential Causes:** * Incomplete clearing of `localStorage` keys (`crawlJobId`, `crawlJobStatus`) when starting a new job. * Frontend state (`currentJobId`, `jobStatus`) not being reset correctly in `handleSubmit`. * Backend issue mixing job data (less likely). **Subtasks:** 1. [x] **Analyze State Reset Logic:** Examined `handleSubmit` in `app/page.tsx`. Found that `currentJobId` and `localStorage` are cleared, but the `selectedUrls` state variable is *not* reset. 2. [ ] **Analyze Polling Logic:** Verify the `useEffect` hook responsible for polling (`/api/crawl-status/[job_id]`) correctly stops polling for the old job ID and uses the new one. 3. [x] **Propose Fix:** Add `setSelectedUrls(new Set());` to the state resets within `handleSubmit` in `app/page.tsx`. 4. [x] **Implement Fix:** Added `setJobStatus(null)` and `setSelectedUrls(new Set())` to `handleSubmit` in `app/page.tsx`. 5. [ ] **Verify Fix:** Test by stopping a crawl, starting a new one for a different domain, and confirming only the new URLs appear. 6. [ ] **Seal Task List:** Mark as complete once verified by the user.

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/cyberagiinc/DevDocs'

If you have feedback or need assistance with the MCP directory API, please join our Discord server