Fix the problem with archive_command and then your WAL files will resume shipping.
If your pg_xlog directory looks like this:
... lots and lots ...
-rw------- 1 postgres postgres 16777216 Jun 22 15:20 00000001000001F200000095
-rw------- 1 postgres postgres 16777216 Jun 22 15:22 00000001000001F200000096
-rw------- 1 postgres postgres 16777216 Jun 22 15:24 00000001000001F200000097
-rw------- 1 postgres postgres 16777216 Jun 22 15:26 00000001000001F200000098
-rw------- 1 postgres postgres 16777216 Jun 22 15:28 00000001000001F200000099
-rw------- 1 postgres postgres 16777216 Jun 22 15:30 00000001000001F20000009A
-rw------- 1 postgres postgres 16777216 Jun 22 15:32 00000001000001F20000009B
-rw------- 1 postgres postgres 16777216 Jun 22 15:34 00000001000001F20000009C
... lots and lots ...
Then something has gone wrong with your WAL file shipping and the WAL files are accumulating on your master db until they can be shipped (to a remote server). Check the log messages in /var/log/postgresql/postgresql-8.4-main.log for any relating to "archive command" to find the problem:
grep "archive command" /var/log/postgresql/postgresql-8.4-main.log
...
2012-06-22 16:12:37 BST LOG:  archive command failed with exit code 1
2012-06-22 16:12:37 BST DETAIL:  The failed archive command was: test -f /mnt/nfs/psql_archive/mounted && test ! -f /mnt/nfs/psql_archive/00000001000001AA0000005B && rsync -a pg_xlog/00000001000001AA0000005B /mnt/nfs/psql_archive/00000001000001AA0000005B
In my case, the mounted marker in an nfs share had gone missing so WAL shipping was halted.