Fix/dashboard polling backoff - #2054
Conversation
utksh1
left a comment
There was a problem hiding this comment.
This dashboard polling change is stacked on the unrelated AI-summary, reporting, and Findings edits from #2043. Please rebase onto current main with only Dashboard.tsx plus focused polling/backoff/retry coverage, then rerun checks.
utksh1
left a comment
There was a problem hiding this comment.
This branch combines the already-merged Findings layout change with AI summary behavior and Dashboard polling. Please split it into a focused Dashboard polling PR, remove unrelated files, and rebase before re-review.
92e54ad to
140a63e
Compare
|
Hi @utksh1, addressing your feedback — I found and removed a stray commit that had leaked in an unrelated |
Description
Dashboard.tsx was polling the backend every 10 seconds indefinitely, even when the backend was completely down. This caused unnecessary network requests, stale summary data persisting silently, and no way to recover without a full page reload.
Fixed by adding a
healthFailedstate flag that stops polling immediately after the first health check failure. The Retry Connection button now resets this flag and restarts polling cleanly instead of reloading the entire page.Related Issues
Closes #1852
Type of Change
How Has This Been Tested?
npm run test— all passedChecklist