I have a PostgreSQL server that is not recycling WAL files. Log files are continually created and no old log files are ever removed.
Running PostgreSQL v 9.4.8 with archive settings:
archive_mode = on
archive_command = /bin/true
Checkpoint logging is enabled and does not record any logs being recycled:
2016-05-16 00:05:37 EDT - LOG: checkpoint starting: xlog
--
2016-05-16 00:09:51 EDT - LOG: checkpoint complete: wrote 38574 buffers (29.4%); 0 transaction log file(s) added, 0 removed, 0 recycled; write=252.921 s, sync=1.340 s, total=254.312 s; sync files=48, longest=1.101 s, average=0.027 s
2016-05-16 00:10:37 EDT - LOG: checkpoint starting: time
2016-05-16 00:14:11 EDT - LOG: checkpoint complete: wrote 43786 buffers (33.4%); 0 transaction log file(s) added, 0 removed, 0 recycled; write=206.271 s, sync=7.428 s, total=213.765 s; sync files=82, longest=3.821 s, average=0.090 s
...
2016-05-16 07:32:27 EDT - LOG: checkpoint complete: wrote 7617 buffers (5.8%); 1 transaction log file(s) added, 0 removed, 0 recycled; write=269.3
Control data records:
pg_control last modified: Mon 16 May 2016 06:23:07 PM EDT
Latest checkpoint location: 1EB/1F008FE8
Prior checkpoint location: 1EB/1E0009E0
Latest checkpoint's REDO location: 1EB/1F008FB0
There is a single replication client:
pid | 23287
usesysid | 10
usename | postgres
application_name | walreceiver
client_addr | 172.16.56.246
client_hostname |
client_port | 58070
backend_start | 2016-05-14 03:27:33.523611+00
backend_xmin |
state | streaming
sent_location | 1EB/1F042760
write_location | 1EB/1F042760
flush_location | 1EB/1F042760
replay_location | 1EB/1F037F40
sync_priority | 0
sync_state | async
Does anyone have suggestions what else to check? What can cause WALs not to be recycled or removed?
Thanks in advance.
Scott Moynes
VM Farms
