Why checkpoint_timeout had maximum value of 1h?

Hi all,

Is there any real reason why checkpoint_timeout limited to 1hour?

In my case I have some replicas with WAL on SAS raid and PGDATA on SSD with
limited write endurance.
And I don't worry about possible long time recovery after power failure in
that case.
Whats more working dataset fill in shared buffers, so almost no dirty
buffers evictions by bgwriter or backends happened.

In that case having checkpoint_timeout=10hour could reduce amout of writes
on SSD by factor of 10, and increase planned ssd lifetime by the same
amount.

I would like to have ability to set checkpoint_timeout=high value
and (whats even better) checkpoint_timeout=0 - in that case checkpoint
happen when all checkpoint_segments were used.

Is there any serious drawbacks in that idea?

Is it safe to increase that limit in source and rebuild database? (9.0 and
9.1 case)

--
Maxim Boguk
Senior Postgresql DBA.

Phone RU: +7 910 405 4718
Phone AU: +61 45 218 5678

Skype: maxim.boguk
Jabber: maxim(dot)boguk(at)gmail(dot)com

LinkedIn profile: http://nz.linkedin.com/in/maximboguk
"If they can send one man to the moon... why can't they send them all?"

МойКруг: http://mboguk.moikrug.ru/
"People problems are solved with people.
If people cannot solve the problem, try technology.
People will then wish they'd listened at the first stage."