Re: pg_xlog size growing untill it fills the partition

Hello,

I have a problem on my pg 9.2.4 setup (OpenSuse 12.2, kernel 3.2.13).
My pg_xlog directory is growing uncontrolably untill it fills the partition. The database is under heavy write load and is spread on two tablesapces one on a ssd software raid1 partition and a second one on a hdd software raid1 partition.
I have no wal archiving enabled nor any replication.

I have tried different checkpoint related parameters without any noticable improvement.
Now I have:
        checkpoint_completion_target = 0.9
        wal_buffers = 8MB
        checkpoint_segments = 16
        checkpoint_timeout = 20min
        shared_buffers = 2GB
        log_checkpoints = on

This is what I can see in the log:
2013-10-03 13:58:56 CEST   LOG:  checkpoint starting: xlog
2013-10-03 13:59:56 CEST   LOG:  checkpoint complete: wrote 448 buffers (0.2%); 0 transaction log file(s) added, 9 removed, 18 recycled; write=39.144 s, sync=21.136 s, total=60.286 s; sync files=380, longest=14.517 s, average=0.055 s
2013-10-03 14:04:07 CEST   LOG:  checkpoint starting: xlog
2013-10-03 15:27:01 CEST   LOG:  checkpoint complete: wrote 693 buffers (0.3%); 0 transaction log file(s) added, 0 removed, 16 recycled; write=90.775 s, sync=4883.295 s, total=4974.074 s; sync files=531, longest=152.855 s, average=9.196 s
2013-10-03 15:27:01 CEST   LOG:  checkpoint starting: xlog time
2013-10-03 19:06:30 CEST   LOG:  checkpoint complete: wrote 3467 buffers (1.3%); 0 transaction log file(s) added, 0 removed, 16 recycled; write=122.555 s, sync=13046.077 s, total=13168.637 s; sync files=650, longest=234.697 s, average=20.069 s
2013-10-03 19:06:30 CEST   LOG:  checkpoint starting: xlog time
2013-10-03 22:30:25 CEST   LOG:  checkpoint complete: wrote 10198 buffers (3.9%); 0 transaction log file(s) added, 216 removed, 33 recycled; write=132.229 s, sync=12102.311 s, total=12234.608 s; sync files=667, longest=181.374 s, average=18.144 s
2013-10-03 22:30:25 CEST   LOG:  checkpoint starting: xlog time

I ran pg_test_fsync on the hard disk when postgresql is down and I have the following results:
-------------------------------------------------------------------------------------------------------------
O_DIRECT supported on this platform for open_datasync and open_sync.

Compare file sync methods using one 8kB write:
(in wal_sync_method preference order, except fdatasync
is Linux's default)
        open_datasync                      59.639 ops/sec
        fdatasync                          43.959 ops/sec
        fsync                              34.497 ops/sec
        fsync_writethrough                            n/a
        open_sync                          35.612 ops/sec

Compare file sync methods using two 8kB writes:
(in wal_sync_method preference order, except fdatasync
is Linux's default)
        open_datasync                      24.199 ops/sec
        fdatasync                          38.680 ops/sec
        fsync                              35.412 ops/sec
        fsync_writethrough                            n/a
        open_sync                          13.337 ops/sec

Compare open_sync with different write sizes:
(This is designed to compare the cost of writing 16kB
in different write open_sync sizes.)
         1 * 16kB open_sync write           5.499 ops/sec
         2 *  8kB open_sync writes         21.412 ops/sec
         4 *  4kB open_sync writes         13.065 ops/sec
         8 *  2kB open_sync writes          6.720 ops/sec
        16 *  1kB open_sync writes          3.320 ops/sec

Test if fsync on non-write file descriptor is honored:
(If the times are similar, fsync() can sync data written
on a different descriptor.)
        write, fsync, close                36.353 ops/sec
        write, close, fsync                37.347 ops/sec

Non-Sync'ed 8kB writes:
        write                           500365.249 ops/sec
-------------------------------------------------------------------------------------------------------------

When the server is up and running under the usual load I get the following results:
-------------------------------------------------------------------------------------------------------------
2 seconds per test
O_DIRECT supported on this platform for open_datasync and open_sync.

Compare file sync methods using one 8kB write:
(in wal_sync_method preference order, except fdatasync
is Linux's default)
        open_datasync                       0.369 ops/sec
        fdatasync                           0.575 ops/sec
        fsync                               0.125 ops/sec
        fsync_writethrough                            n/a
        open_sync                           0.222 ops/sec

Compare file sync methods using two 8kB writes:
(in wal_sync_method preference order, except fdatasync
is Linux's default)
        open_datasync                       0.383 ops/sec
        fdatasync                           2.171 ops/sec
        fsync                               1.318 ops/sec
        fsync_writethrough                            n/a
        open_sync                           0.929 ops/sec

Compare open_sync with different write sizes:
(This is designed to compare the cost of writing 16kB
in different write open_sync sizes.)
         1 * 16kB open_sync write           0.079 ops/sec
         2 *  8kB open_sync writes          0.041 ops/sec
         4 *  4kB open_sync writes          0.194 ops/sec
         8 *  2kB open_sync writes          0.013 ops/sec
        16 *  1kB open_sync writes          0.005 ops/sec

Test if fsync on non-write file descriptor is honored:
(If the times are similar, fsync() can sync data written
on a different descriptor.)
        write, fsync, close                 0.098 ops/sec
        write, close, fsync                 0.067 ops/sec

Non-Sync'ed 8kB writes:
        write                               0.102 ops/sec
-------------------------------------------------------------------------------------------------------------

I need to tell to the server to limit the amount of wal files in pg_xlog somehow whatever the efect on the performance could be.

As for now I have to monitor the disk size and manually restart the server. It takes a few seconds to process all the wal files and resume.
I have to do this a few times a day as it takes about 10 hours to get 20GB of wal files in the pg_xlog directory (what fills my partion on the ssd).

Any help with this issue would be greatly appreciated as I don't see what else I can try.

Michal


--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general