Add more TAP coverage for archive status with crash recovery of standbys

This part of the test was included originally in 4e87c48, but got
reverted as of f9c1b8d because of timing issues in the test, where,
after more analysis, we found that the standbys may not have recovered
from the archives all the segments needed by the test.  This stabilizes
the test by making sure that standbys replay up to the position returned
by pg_switch_wal(), meaning that all segments are recovered before the
manual checkpoint that tests WAL segment recycling and its interactions
with archive status files.

Author: Jehan-Guillaume de Rorthais
Reviewed-by: Kyotaro Horiguchi, Michael Paquier
Discussion: https://postgr.es/m/[email protected]

Author: michaelpq

src/test/recovery/t/020_archive_status.pl

@@ -5,7 +5,7 @@
 use warnings;
 use PostgresNode;
 use TestLib;
-use Test::More tests => 13;
+use Test::More tests => 16;
 use Config;
 
 my $primary = get_new_node('master');
@@ -119,10 +119,16 @@
 $primary->safe_psql(
 	'postgres', q{
 	INSERT INTO mine SELECT generate_series(10,20) AS x;
-	SELECT pg_switch_wal();
 	CHECKPOINT;
 });
 
+# Switch to a new segment and use the returned LSN to make sure that
+# standbys have caught up to this point.
+my $primary_lsn = $primary->safe_psql(
+	'postgres', q{
+	SELECT pg_switch_wal();
+});
+
 $primary->poll_query_until('postgres',
 	q{ SELECT last_archived_wal FROM pg_stat_archiver },
 	$segment_name_2)
@@ -134,14 +140,32 @@
 $standby1->append_conf('postgresql.conf', "archive_mode = on");
 my $standby1_data = $standby1->data_dir;
 $standby1->start;
+
+# Wait for the replay of the segment switch done previously, ensuring
+# that all segments needed are restored from the archives.
+$standby1->poll_query_until('postgres',
+	qq{ SELECT pg_wal_lsn_diff(pg_last_wal_replay_lsn(), '$primary_lsn') >= 0 }
+) or die "Timed out while waiting for xlog replay on standby2";
+
 $standby1->safe_psql('postgres', q{CHECKPOINT});
 
+# Recovery with archive_mode=on does not keep .ready signal files inherited
+# from backup.  Note that this WAL segment existed in the backup.
+ok( !-f "$standby1_data/$segment_path_1_ready",
+	".ready file for WAL segment $segment_name_1 present in backup got removed with archive_mode=on on standby"
+);
+
 # Recovery with archive_mode=on should not create .ready files.
 # Note that this segment did not exist in the backup.
 ok( !-f "$standby1_data/$segment_path_2_ready",
 	".ready file for WAL segment $segment_name_2 not created on standby when archive_mode=on on standby"
 );
 
+# Recovery with archive_mode = on creates .done files.
+ok( -f "$standby1_data/$segment_path_2_done",
+	".done file for WAL segment $segment_name_2 created when archive_mode=on on standby"
+);
+
 # Test recovery with archive_mode = always, which should always keep
 # .ready files if archiving is enabled, though here we want the archive
 # command to fail to persist the .ready files.  Note that this node
@@ -153,12 +177,22 @@
 my $standby2_data = $standby2->data_dir;
 $standby2->start;
 
+# Wait for the replay of the segment switch done previously, ensuring
+# that all segments needed are restored from the archives.
+$standby2->poll_query_until('postgres',
+	qq{ SELECT pg_wal_lsn_diff(pg_last_wal_replay_lsn(), '$primary_lsn') >= 0 }
+) or die "Timed out while waiting for xlog replay on standby2";
+
 $standby2->safe_psql('postgres', q{CHECKPOINT});
 
 ok( -f "$standby2_data/$segment_path_1_ready",
 	".ready file for WAL segment $segment_name_1 existing in backup is kept with archive_mode=always on standby"
 );
 
+ok( -f "$standby2_data/$segment_path_2_ready",
+	".ready file for WAL segment $segment_name_2 created with archive_mode=always on standby"
+);
+
 # Reset statistics of the archiver for the next checks.
 $standby2->safe_psql('postgres', q{SELECT pg_stat_reset_shared('archiver')});