Re: Proposal: Conflict log history table for Logical Replication - Mailing list pgsql-hackers

From Zsolt Parragi
Subject Re: Proposal: Conflict log history table for Logical Replication
Date
Msg-id CAN4CZFOSRonwrbyLi38dfcYXe3k1gaBXOWSELjLYOXFdjG8_+w@mail.gmail.com
Whole thread
List pgsql-hackers
Hello

+                snprintf(conflictlogtable,
+                         sizeof(conflictlogtable),
+                         "%s.pg_conflict_log_for_subid_%s",
+                         conflict_schema, subid);

This and a few some places / commit message use
pg_conflict_log_for_subid_<oid>, but the proper name seems to be

+    snprintf(relname, NAMEDATALEN, "pg_conflict_log_%u", subid);

@@ -1767,6 +1771,10 @@ apply_handle_stream_start(StringInfo s)
                 (errcode(ERRCODE_PROTOCOL_VIOLATION),
                  errmsg_internal("invalid transaction ID in streamed replication
transaction")));

+    remote_xid = stream_xid;
+    remote_final_lsn = InvalidXLogRecPtr;
+    remote_commit_ts = 0;

Shouldn't apply_handle_begin_prepare also write out the xid/commit_ts?



pgsql-hackers by date:

Previous
From: Bertrand Drouvot
Date:
Subject: Add per-backend lock statistics
Next
From: Zsolt Parragi
Date:
Subject: Re: alert clients when prepared statements are deallocated