Add a note to the documentation of pg_standby that it's important that

the postgres process has permissions to delete the trigger file, per
suggestion by Mason Hale.

Also fix pg_standby to do a more predictable exit(200) instead of the
current exit(-1) when the unlink of the trigger file fails anyway.

This only affects 8.3 branch. Older versions didn't have pg_standby,
and in 8.4 upwards pg_standby is no longer responsible for deleting the
trigger file; it's supposed to be done by recovery_end_command instead.

0 files changed, 0 insertions, 0 deletions