Strange memory leak
rra at stanford.edu
Tue Dec 21 07:49:35 UTC 2004
Andrew MacLennan <apm at connect.com.au> writes:
> A little over two months ago I upgraded our news architecture to newer
> hardware (2 x Sun V240s, 2Gb RAM, Solaris 8) and INN 2.4.1
> After cutting them across to live operation we noticed a memory leak,
> chewing up memory through the kernel and not releasing it. A reboot
> once a week was the only way to free it up again (and hardly an elegant
What are you using for storage? This sounds exactly like a problem that
we had, and Sun support, after being mostly useless for a year, finally
managed to figure out that their A1000 RAID drivers had a bug involving
memory mapping that required an upgrade. The upgraded version didn't
support memory mapping of raw devices any more, so we had to reconfigure
the server to use disk files instead.
This is a kernel memory leak, right? As in, the memory doesn't show up on
top and none of the processes are consuming much in the way of memory, but
the system slowly starts swapping more and more and eventually expire
takes half of forever to finish and the system has to be rebooted?
Russ Allbery (rra at stanford.edu) <http://www.eyrie.org/~eagle/>
Please send questions to the list rather than mailing me directly.
<http://www.eyrie.org/~eagle/faqs/questions.html> explains why.
More information about the inn-workers