On Mon, Nov 16, 2026 at 1:39 PM Zhijie Hou <[email protected]> wrote: > > Attached is the v3 patch set (with the TAP test split into 0002). Please > let me know what you think. >
* @@ -335,10 +345,10 @@ update_local_synced_slot(RemoteSlot *remote_slot, Oid remote_dbid) SpinLockRelease(&slot->mutex); updated_xmin_or_lsn = true; + found_consistent_snapshot = true; } This change makes me wonder whether found_consistent_point is a good name for this variable as in this loop, we didn’t find anything. How about consistent_snapshot_available or has_consistent_snapshot? I have used the earlier in the attached and slightly changed the comments. -- With Regards, Amit Kapila.
v4-0001-Fix-premature-persistence-of-synced-slots-without.patch
Description: Binary data
