postgreSQL archive process failed on

Question

I upgraded a cluster from 8.3 to 8.4. Things seemed to be fine so I dropped the 8.3 cluster. Then noticed an error and found that the archive_command was still pointing to the 8.3 data directory

archive_command = 'cp "%p" /var/lib/postgresql/8.3/main/wal_archives/"%f"'

I changed this to 8.4 but now continually get the following errors in the log file and process list for one specific file

2014-10-09 17:02:12 CDT DETAIL:  The failed archive command was: cp "pg_xlog/000000010000000000000012" /var/lib/postgresql/8.4/main/wal_archives/"000000010000000000000012"
cp: cannot create regular file `/var/lib/postgresql/8.4/main/wal_archives/000000010000000000000012': No such file or directory
 4122 ?        Ss     0:00  \_ postgres: archiver process   failed on 000000010000000000000012

I'm not sure the best way to recover from this. As far as I can tell the database is fully functional

Well, does `/var/lib/postgresql/8.4/main/wal_archives/` exist? — Craig Ringer, Oct 10 '14 at 02:33
It does now. Seems so obvious now that you said it. I think your answer. "Make sure the wal_archive directory exits" is perfectly valid. — Dustin Butler, Oct 10 '14 at 15:10

score 0 · Accepted Answer · answered Oct 11 '14 at 02:48

0

Check to ensure that /var/lib/postgresql/8.4/main/wal_archives exists.

You will also want to ensure it's writeable by the postgres user, though if this was a permissions problem the error would be different.

answered Oct 11 '14 at 02:48

Craig Ringer

11,083
9
40
61

score 0 · Answer 2 · answered Nov 17 '14 at 11:23

Check that the user postgres is owner of this dir (/var/lib/postgresql/8.4/main/wal_archives) or he can write to this! (chwon/chmod) Also it is better that you don't you this path /var/lib/postgresql/8.4/main/ better you use an other device or storage.

postgreSQL archive process failed on

2 Answers2