Skip to content

Fix Cloud Server setup leaving the node without the drive - #1919

Closed
joepio wants to merge 1 commit into
developfrom
fix/cloud-server-sync-flow
Closed

joepio wants to merge 1 commit into
developfrom
fix/cloud-server-sync-flow

Conversation

@joepio

@joepio joepio commented Sep 29, 2026

Copy link
Copy Markdown
Member

Requested by Joep

Summary

Cloud Server setup could report "connected" while the node never received the drive, so every later commit (chat message, table row) was refused with Parent of … not found, retried over HTTP, and retried forever (endless spinner, warning icons, card never blue).

  • Setup now runs the version-vector reconcile against the node and verifies the node's inventory covers the local drive (root included) before reporting connected. This also covers drives that were already synced elsewhere, where switching servers copied nothing. New Store.syncDriveToServerAndVerify.
  • Parent of … not found is classified as a blocking refusal, so it is parked with its reason and the edit is kept, instead of being resent over HTTP.
  • Sync page request storm: the 5s poll no longer creates a new object each tick, the enrollment effect depends on booleans, and /api/me on window focus is throttled.
  • Avatar/file object URLs are cached (bounded LRU) so revisiting settings is instant.
  • Store errors now reach Sentry. They previously only went to Bugsnag, which never fired because !isDev tested the function instead of calling it.

Not in this PR

  • Stuck-sync UI (replace the endless spinner with reason plus Retry) and pushing missing parents from a blocked entry.
  • Making sign-out on a managed build skip the FOSS welcome screen.
  • Restricting the WS-to-HTTP fallback to true transport failures.

Test plan

  • local-outbox and cloudSync unit tests updated/added; full browser/lib suite passes
  • pnpm lint clean
  • Manual: enable Cloud Server on an existing synced drive, confirm the card turns blue and posting a message succeeds

Setup switched servers without copying an already-synced drive and reported
success without waiting for the reconcile, so the node lacked the drive root
and every commit was refused as "Parent of ... not found", retried forever.

- Sync the drive to the node and verify its inventory before reporting connected
- Park "Parent ... not found" as a blocking refusal instead of retrying over HTTP
- Stop the sync page request storm (compare polled info, narrow effect deps, throttle /api/me)
- Cache avatar object URLs so remounts are instant
- Send store errors to Sentry and fix the Bugsnag isDev gate that never fired

joepio commented Oct 5, 2026

Copy link
Copy Markdown
Member Author

Landed on develop in the combined batch #2060 (95707c0), tested together in one full run. Closing as included.


Generated by Claude Code

@joepio joepio closed this Oct 5, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant