Feature guide · 3 minute read
Clear and rerun a crawl
Start with fresh results after a site change or a revised crawl setup. Keep the evidence you need first, then clear the current run and check the affected pages again.
Best for post-fix checks, repeatable audits, and changes to crawl scope
20-second workflow
Start fresh. Check the same issue again.
The same public fixture folder returns 14 pages again. Its deliberate broken-target URL still returns 404; a fresh crawl verifies the current state, not a repair.
What happens in the demo
Request a reset
Select Clear in the completed fixture crawl. The confirmation lets you cancel before removing the current results.
Confirm and check the empty table
Confirm Clear. The results disappear, but the fixture URL and folder scope remain available.
Run the same folder again
Select Crawl folder. The recording keeps the crawl in progress through completion, without skipping its loading states.
Return to the affected URL
Open 4xx Client Errors, then select the broken-target URL. The intentional fixture still returns 404: rerunning the crawl did not change the website.
Keep the evidence before clearing it.
Clear removes the current page, link, and issue data. Export anything you need for a comparison or handoff before confirming. An empty table is not a saved copy of the previous run.
Record the starting URL, crawl scope, relevant settings, and the specific URLs you plan to verify. A different starting scope can change which pages are discovered, even when the website has not changed.
Clear, check the scope, and crawl again.
- Wait for the current crawl to finish. If it is still running, pause it before attempting to clear it.
- Select Clear in the toolbar. Read the confirmation dialog, then choose Clear only when you no longer need the current results. Choose Cancel to keep them.
- Check the starting URL and scope in the now-empty crawl view. Keep them the same for a comparable rerun, or change them deliberately for a different investigation.
- Start the crawl with the main crawl button. Its label reflects the selected scope, such as Crawl folder for a subfolder crawl.
- Wait for completion before judging the result. Return to the relevant issue group and inspect the affected URL, not just the overall issue count.
Practice with the public fixture crawl before using this sequence on an audit whose evidence you need to retain.
Clear and New crawl share a confirmation.
- Clear
- Removes the current results after confirmation. It does not start another crawl automatically. The starting URL remains available so you can review it before running again.
- New crawl
- After a crawl completes, the main button offers New crawl. It asks to clear the existing results first and then focuses the URL field. Review the input and start the next crawl separately.
- Pause and resume
- Pause keeps an unfinished run available to resume. Clearing a paused run stops it and removes its data; that is different from continuing where it left off.
A fresh crawl is evidence, not a fix.
Check that the page you care about was actually reached. A smaller issue count can mean a problem was fixed, but it can also mean the new scope, crawl settings, or discovery path excluded that page.
- For a reported 404, inspect the new response for that URL and the source page linking to it.
- For a metadata issue, read the current field value beside the page URL.
- If the expected page is missing, check scope and active table filters before treating the issue as resolved.
Use the earlier export as your baseline. This workflow produces a new run; it does not provide an automatic before-and-after comparison.