You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit fd71df5
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: journeys/paykit-clock-changes.md
+3Lines changed: 3 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -38,5 +38,8 @@ Network fault injection is not provided by the journey capability table. Use a d
38
38
3. Re-enable connectivity while leaving Bitkit open. Verify that the same identity and contact list recover without signing out or restarting the app when the saved session or adopted Ring credential is still available. If the Ring credential is unavailable, select the same identity again from the normal identity choice flow.
39
39
4. Repeat the failed startup and restore connectivity while Bitkit is backgrounded. Return to the foreground from a profile/contact screen and verify the same recovery. Resume must work from any screen, not only Home.
40
40
5. Start adopting a Ring identity while recovery is pending. Verify that automatic restoration does not replace the selected identity. Explicit sign-out or wallet reset must not be undone by a pending restoration.
41
+
6. While online, use an isolated shared-state lock holder to make a saved-session restore return temporarily unavailable. Keep Bitkit in the foreground without navigating, reconnecting, or resuming it. Release the held lock during the short retry window and verify recovery without another action. Android makes up to eight retries with a five-second delay between completed attempts; SDK work and safety waits can take additional time. After this window, the existing maintenance and foreground/network triggers remain available. Backgrounding the app or disabling Paykit must stop queued retries, without interrupting an active SDK write. Repeat with a permanent credentials failure and verify it does not enter this short retry loop.
41
42
42
43
Both platforms retry automatically on connectivity restoration and app resume. A valid saved session must recover without a new authorization. Android can also recover with an available adopted Ring credential.
44
+
45
+
The controlled lock-holder check needs a separate fixture and is not established by the unit tests for retry scheduling.
0 commit comments