mirror of
https://github.com/stablyai/orca.git
synced 2026-10-19 00:03:20 +00:00
On wake, remote/SSH runtimes reconnect in a staggered burst; the sidebar refetched all worktrees once per host, piling up K synchronous full-sidebar remounts and freezing the UI. Wrap the reconnect refresh in a single-flight coalescer so at most one refresh runs at a time plus one queued rerun, for any K. fetchAllWorktrees and its lineage follow-up are unchanged; only how often they fire changes. Co-authored-by: Orca <help@stably.ai>