When ever master node performs vacuum, he also logs the snapshotConflictHorizon for few specific vacuum records for example
rmgr: Heap2 len (rec/tot): 64/ 64, tx: 0, lsn: 1/D8061CF0, prev 1/D8061C68, desc: PRUNE_VACUUM_SCAN snapshotConflictHorizon: 3837, isCatalogRel: T, nplans: 0, nredirected: 1, ndead: 0, nunused: 2, redirected: [1->7], unused: [3, 6], blkref #0: rel 1663/5/16926 blk 0
rmgr: Heap2 len (rec/tot): 134/ 134, tx: 3835, lsn: 1/D80558E8, prev 1/D8055898, desc: PRUNE_ON_ACCESS snapshotConflictHorizon: 3833, isCatalogRel: T, nplans: 0, nredirected: 0, ndead: 40, nunused: 0, dead: [13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65], blkref #0: rel 1663/5/16921 blk 55
Now standby tries to replay those WALs and invalidates the replication slot. Im curious to know the fastest way to get the last applied snapshotConflictHorizon transactionId.Is there any PostgreSQL system table where this data is stored?
