> Quoting Jukka Marin,
> 
> > I think this problem is real, considering the number of people who have
> > reported it.
> 
> It certainly is real. It's an old problem in the VM code. If I remember
> correctly, it has to do with VM objects not being cleaned up when having
> a pager associated with them, and it occurs in situations where a
> child process exits while pages in the parent have been paged out.

I am running one NetBSD -current (Sep 5) machine as a router/terminal server. It had to be rebooted regularly because of out-of swap conditions. 

In this case, cron was causing the swap leakage. It had only one empty crontab for root, so it was only scanning the file, not forking and execing. When cron was running, killing it would return all the lost page to the available swap.

I have removed cron and the machine has been up for 31 days now, with not a page of swap leakage.


-- 
Andreas Wrede                  Planix, Inc.
andreas@planix.com             Networking, System Administration, Consulting
http://www.planix.com          Toronto, Ontario, Canada