|
75 | 75 | # |
76 | 76 | $node_1->stop('fast'); |
77 | 77 | $node_3->promote; |
| 78 | +# Force a checkpoint after the promotion. pg_rewind looks at the control |
| 79 | +# file to determine what timeline the server is on, and that isn't updated |
| 80 | +# immediately at promotion, but only at the next checkpoint. When running |
| 81 | +# pg_rewind in remote mode, it's possible that we complete the test steps |
| 82 | +# after promotion so quickly that when pg_rewind runs, the standby has not |
| 83 | +# performed a checkpoint after promotion yet. |
| 84 | +$node_3->safe_psql('postgres', "checkpoint"); |
78 | 85 |
|
79 | 86 | # reconfigure node_1 as a standby following node_3 |
80 | 87 | my $node_3_connstr = $node_3->connstr; |
|
99 | 106 | $node_3->wait_for_catchup('node_1', 'replay', $lsn); |
100 | 107 |
|
101 | 108 | $node_1->promote; |
| 109 | +# Force a checkpoint after promotion, like earlier. |
| 110 | +$node_1->safe_psql('postgres', "checkpoint"); |
102 | 111 |
|
103 | 112 | # |
104 | 113 | # We now have a split-brain with two primaries. Insert a row on both to |
105 | 114 | # demonstratively create a split brain. After the rewind, we should only |
106 | 115 | # see the insert on 1, as the insert on node 3 is rewound away. |
107 | 116 | # |
108 | 117 | $node_1->safe_psql('postgres', "INSERT INTO public.foo (t) VALUES ('keep this')"); |
| 118 | +# 'bar' is unmodified in node 1, so it won't be overwritten by replaying the |
| 119 | +# WAL from node 1. |
| 120 | +$node_3->safe_psql('postgres', "INSERT INTO public.bar (t) VALUES ('rewind this')"); |
109 | 121 |
|
110 | 122 | # Insert more rows in node 1, to bump up the XID counter. Otherwise, if |
111 | 123 | # rewind doesn't correctly rewind the changes made on the other node, |
|
114 | 126 | $node_1->safe_psql('postgres', "INSERT INTO public.foo (t) VALUES ('and this')"); |
115 | 127 | $node_1->safe_psql('postgres', "INSERT INTO public.foo (t) VALUES ('and this too')"); |
116 | 128 |
|
117 | | -# Also insert a row in 'bar' on node 3. It is unmodified in node 1, so it won't get |
118 | | -# overwritten by replaying the WAL from node 1. |
119 | | -$node_3->safe_psql('postgres', "INSERT INTO public.bar (t) VALUES ('rewind this')"); |
120 | | - |
121 | 129 | # Wait for node 2 to catch up |
122 | 130 | $node_2->poll_query_until('postgres', |
123 | 131 | q|SELECT COUNT(*) > 1 FROM public.bar|, 't'); |
|
139 | 147 | [ |
140 | 148 | 'pg_rewind', |
141 | 149 | "--source-server=$node_1_connstr", |
142 | | - "--target-pgdata=$node_2_pgdata" |
| 150 | + "--target-pgdata=$node_2_pgdata", |
| 151 | + "--debug" |
143 | 152 | ], |
144 | | - 'pg_rewind detects rewind needed'); |
| 153 | + 'run pg_rewind'); |
145 | 154 |
|
146 | 155 | # Now move back postgresql.conf with old settings |
147 | 156 | move( |
|
153 | 162 | # Check contents of the test tables after rewind. The rows inserted in node 3 |
154 | 163 | # before rewind should've been overwritten with the data from node 1. |
155 | 164 | my $result; |
156 | | -$result = $node_2->safe_psql('postgres', 'checkpoint'); |
157 | 165 | $result = $node_2->safe_psql('postgres', 'SELECT * FROM public.foo'); |
158 | 166 | is($result, qq(keep this |
159 | 167 | and this |
|
0 commit comments