Skip to content

Reload a drive after restoring it from Cloud Vault - #1791

Closed
joepio wants to merge 1 commit into
developfrom
claude/project-thread-u8vgm0-returning
Closed

joepio wants to merge 1 commit into
developfrom
claude/project-thread-u8vgm0-returning

Conversation

@joepio

@joepio joepio commented Sep 25, 2026 •

Copy link
Copy Markdown
Member

Before: a returning user who signed in on a new device and got their drive back from Cloud Vault often saw the drive's raw identifier (atomic:9y6J…) as its title, or an empty new "My drive" in its place. The app kept the "not found" it got before the restore and trusted that over the data the restore had just written. The atomic-saas e2e returning-user-second-browser.spec.ts failed on this title check in both the node and nodeless jobs.

After: once the restore has written the drive, the app reloads it and its agent from the local database. Sign-in sees the drive, no new home is made over it, and the sidebar shows its name.

How: restoreFromVault calls reopenRestoredDrive after an import that restored anything, like the Sync page restore already does. getResource returns a cached error unchanged, so without the reload the sign-in check at GettingStartedFlow.tsx:710 read the old error and openPrivateHome went on to create "My drive". If that did not happen, later local saves merged the resource into itself and kept the stale error.

Related Issues

No issue. Found through atomic-saas PR 152 CI (the "Portal e2e" jobs).

Checklist

  • Add changelog entry linking to issue, describe API changes
  • Add or update tests if needed
  • Update docs if needed

Validation: on a local stack running develop 87301c5, the title check failed in 5 of 8 node-mode runs and 1 of 3 nodeless runs. With this commit it passed 4 of 4 node runs and 3 of 3 nodeless runs. Workspace lint, data-browser tsc --noEmit and the vitest suites for helpers/managed and views/getting-started (302 tests) pass.

A vault restore writes the drive into the local database only. The store
still held the lookup from before the restore: a "not found" from the
node, or "not available locally" without one. `getResource` returns a
cached error as is, so sign-in's check right after the restore read the
old answer, concluded the device had nothing, and let `openPrivateHome`
create a new home over the restored drive. When it did not, the drive
kept the stale error through every later save, and the sidebar showed
its raw subject instead of its name.

`restoreFromVault` now reloads the drive and its agent once the import
has written something, local database first, the way the Sync page
restore already did through `reopenRestoredDrive`.
joepio added a commit that referenced this pull request Sep 25, 2026
# Conflicts:
#	browser/CHANGELOG.md
joepio added a commit that referenced this pull request Sep 25, 2026
#1826, #1787, #1783, #1791, #1840, #1841) (#1844)

Combined batch, green on Mancave in run 4645 on 3346a6c (CI and Downstream).

Includes #1826, #1787, #1783, #1791, #1840 and #1841, plus settled
translation catalogs.

joepio commented Sep 25, 2026

Copy link
Copy Markdown
Member Author

Landed on develop in ef161c3 via the combined batch #1844 (green on Mancave, run 4645). Closing.


Generated by Claude Code

@joepio joepio closed this Sep 25, 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