rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Improve stability of TAP test for synchronous replication 

Slow buildfarm machines have run into issues with this TAP test caused
by a race condition related to the startup of a set of standbys, where
it is possible to finish with an unexpected order in the WAL sender
array of the primary.

This closes the race condition by making sure that any standby started
is registered into the WAL sender array of the primary before starting
the next one based on lookups of pg_stat_replication.

Backpatch down to 9.6 where the test has been introduced.

Author: Michael Paquier
Reviewed-by: Álvaro Herrera, Noah Misch
Discussion: https://postgr.es/m/20190617055145.GB18917@paquier.xyz
Backpatch-through: 9.6
This commit is contained in:
Michael Paquier 2019-07-24 10:53:39 +09:00
parent 5562272a42
commit 7d81bdc8c0
1 changed files with 33 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

42
src/test/recovery/t/007_sync_rep.pl
View File

@ -27,6 +27,23 @@ sub test_sync_state
return;
}
# Start a standby and check that it is registered within the WAL sender
# array of the given primary. This polls the primary's pg_stat_replication
# until the standby is confirmed as registered.
sub start_standby_and_wait
{
my ($master, $standby) = @_;
my $master_name = $master->name;
my $standby_name = $standby->name;
my $query =
"SELECT count(1) = 1 FROM pg_stat_replication WHERE application_name = '$standby_name'";
$standby->start;
print("### Waiting for standby \"$standby_name\" on \"$master_name\"\n");
$master->poll_query_until('postgres', $query);
}
# Initialize master node
my $node_master = get_new_node('master');
$node_master->init(allows_streaming => 1);
@ -36,23 +53,26 @@ my $backup_name = 'master_backup';
# Take backup
$node_master->backup($backup_name);
# Create all the standbys. Their status on the primary is checked to ensure
# the ordering of each one of them in the WAL sender array of the primary.
# Create standby1 linking to master
my $node_standby_1 = get_new_node('standby1');
$node_standby_1->init_from_backup($node_master, $backup_name,
has_streaming => 1);
$node_standby_1->start;
start_standby_and_wait($node_master, $node_standby_1);
# Create standby2 linking to master
my $node_standby_2 = get_new_node('standby2');
$node_standby_2->init_from_backup($node_master, $backup_name,
has_streaming => 1);
$node_standby_2->start;
start_standby_and_wait($node_master, $node_standby_2);
# Create standby3 linking to master
my $node_standby_3 = get_new_node('standby3');
$node_standby_3->init_from_backup($node_master, $backup_name,
has_streaming => 1);
$node_standby_3->start;
start_standby_and_wait($node_master, $node_standby_3);
# Check that sync_state is determined correctly when
# synchronous_standby_names is specified in old syntax.
@ -82,8 +102,10 @@ $node_standby_1->stop;
$node_standby_2->stop;
$node_standby_3->stop;
$node_standby_2->start;
$node_standby_3->start;
# Make sure that each standby reports back to the primary in the wanted
# order.
start_standby_and_wait($node_master, $node_standby_2);
start_standby_and_wait($node_master, $node_standby_3);
# Specify 2 as the number of sync standbys.
# Check that two standbys are in 'sync' state.
@ -94,7 +116,7 @@ standby3|3|sync),
'2(standby1,standby2,standby3)');
# Start standby1
$node_standby_1->start;
start_standby_and_wait($node_master, $node_standby_1);
# Create standby4 linking to master
my $node_standby_4 = get_new_node('standby4');
@ -126,14 +148,16 @@ standby4|1|sync),
# The setting that * comes before another standby name is acceptable
# but does not make sense in most cases. Check that sync_state is
# chosen properly even in case of that setting.
# The priority of standby2 should be 2 because it matches * first.
# chosen properly even in case of that setting. standby1 is selected
# as synchronous as it has the highest priority, and is followed by a
# second standby listed first in the WAL sender array, which is
# standby2 in this case.
test_sync_state(
$node_master, qq(standby1|1|sync
standby2|2|sync
standby3|2|potential
standby4|2|potential),
'asterisk comes before another standby name',
'asterisk before another standby name',
'2(standby1,*,standby2)');
# Check that the setting of '2(*)' chooses standby2 and standby3 that are stored