On Fri, May 5, 2023 at 5:36 PM Drouvot, Bertrand <bertranddrouvot...@gmail.com> wrote: > > On 5/5/23 12:58 PM, Amit Kapila wrote: > > On Fri, May 5, 2023 at 4:02 PM Drouvot, Bertrand > > <bertranddrouvot...@gmail.com> wrote: > > > How did you concluded that 000000010000000000000003 is the file the > > test is expecting to be removed? > > > because I added a note in the test that way: > > " > @@ -535,6 +539,7 @@ $node_standby->safe_psql('postgres', 'checkpoint;'); > > # Verify that the WAL file has not been retained on the standby > my $standby_walfile = $node_standby->data_dir . '/pg_wal/' . $walfile_name; > +note "BDT WAL file is $standby_walfile"; > ok(!-f "$standby_walfile", > "invalidated logical slots do not lead to retaining WAL"); > " > > so that I can check in the test log file: > > grep "WAL file is" > ./build/testrun/recovery/035_standby_logical_decoding/log/regress_log_035_standby_logical_decoding > [08:23:31.931](2.217s) # BDT WAL file is > /Users/admin/pgsql/build/testrun/recovery/035_standby_logical_decoding/data/t_035_standby_logical_decoding_standby_data/pgdata/pg_wal/000000010000000000000003 >
It seems due to some reason the current wal file is not switched due to some reason. I think we need to add more DEBUG info to find that out. Can you please try to print 'RedoRecPtr', '_logSegNo', and recptr? /* * Delete old log files, those no longer needed for last checkpoint to * prevent the disk holding the xlog from growing full. */ XLByteToSeg(RedoRecPtr, _logSegNo, wal_segment_size); KeepLogSeg(recptr, &_logSegNo); -- With Regards, Amit Kapila.