v6-0004-Handle-overwriting-of-replication-slot-statistic-.patch

From 740fc090763da20ba7988683d5481feed35b6545 Mon Sep 17 00:00:00 2001
From: vignesh <vignesh21@gmail.com>
Date: Sat, 10 Apr 2021 11:33:01 +0530
Subject: [PATCH v6 4/5] Handle overwriting of replication slot statistic
 issue.

There is a remote scenario where one of the replication slots is dropped and
the drop slot statistics message is not received by the statistic collector
process, now if the max_replication_slots is reduced to the actual number of
replication slots that are in use and the server is re-started then the
statistics process will not be aware of this and the statistic collector
process will write beyond the slots available, fixed it by writing the
replication slot count in statistics file and increasing
max_replication_slots if required at statup time base on the value
updated in the statistics file.
---
 src/backend/postmaster/pgstat.c | 36 +++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/src/backend/postmaster/pgstat.c b/src/backend/postmaster/pgstat.c
index e1ec7d8b7d..066829234b 100644
--- a/src/backend/postmaster/pgstat.c
+++ b/src/backend/postmaster/pgstat.c
@@ -3741,6 +3741,13 @@ pgstat_write_statsfiles(bool permanent, bool allDbs)
 		(void) rc;				/* we'll check for error with ferror */
 	}
 
+	if (nReplSlotStats > 0)
+	{
+		fputc('N', fpout);
+		rc = fwrite(&nReplSlotStats, sizeof(nReplSlotStats), 1, fpout);
+		(void) rc;				/* we'll check for error with ferror */
+	}
+
 	/*
 	 * Write replication slot stats struct
 	 */
@@ -3960,6 +3967,7 @@ pgstat_read_statsfiles(Oid onlydb, bool permanent, bool deep)
 	bool		found;
 	const char *statfile = permanent ? PGSTAT_STAT_PERMANENT_FILENAME : pgstat_stat_filename;
 	int			i;
+	int			replslotcount;
 
 	/*
 	 * The tables will live in pgStatLocalContext.
@@ -4196,6 +4204,34 @@ pgstat_read_statsfiles(Oid onlydb, bool permanent, bool deep)
 
 				break;
 
+				/*
+				 * 'N'	Indicates the number of replication slot statistics
+				 * present.
+				 */
+			case 'N':
+				if (fread(&replslotcount, 1, sizeof(replslotcount), fpin)
+					!=sizeof(replslotcount))
+				{
+					ereport(pgStatRunningInCollector ? LOG : WARNING,
+							(errmsg("corrupted statistics file \"%s\"",
+									statfile)));
+					goto done;
+				}
+
+				if (replslotcount > max_replication_slots)
+				{
+					max_replication_slots = replslotcount;
+					replSlotStats = repalloc(replSlotStats,
+											 max_replication_slots
+											 * sizeof(PgStat_ReplSlotStats));
+					/*
+					 * Set the same reset timestamp for all replication slots too.
+					 */
+					for (i = 0; i < max_replication_slots; i++)
+						replSlotStats[i].stat_reset_timestamp = globalStats.stat_reset_timestamp;
+				}
+				break;
+
 				/*
 				 * 'R'	A PgStat_ReplSlotStats struct describing a replication
 				 * slot follows.
-- 
2.25.1