120904 Peter Humphrey wrote:
> On Tuesday 04 September 2012 22:00:48 Neil Bothwick wrote:
>> On Tue, 04 Sep 2012 22:31:23 +0200, Volker Armin Hemmann wrote:
>>> If PORTAGE_TMPDIR fills up no biggy, emerge dies, that's it.
>>> But /tmp filled up? Suddenly you will have lots of strange problems.
>>> Don't do it. Spare yourself some headaches.
>> Good point, maybe I should have mentioned I have a 13GB /tmp.
At the moment, after a few hours catching up with the news with FF,
my memory usage is :
total used free shared buffers cached
Mem: 3960 915 3044 0 63 408
-/+ buffers/cache: 443 3516
Swap: 3820 0 3820
Even compiling LO, it doesn't spill into swap anymore.
I assume having PORTAGE_TMPDIR on SSD wb noticeably faster than on HDD,
but how much faster still would it be to have it in memory ?
Memory is cheap & I could buy another 4 GB , it there were a reason.
> Also having 16GB RAM I've limited /tmp to 10GB.
> I wonder whether 13GB would offer any advantage.
> Unlikely, as the only time it gets used in earnest
> is when compiling Firefox, OO and the like.
> Maybe I should just remove the restriction
> and let the kernel optimise its own use of swap and tmpfs.
> This box spends well over 90% of its cycles on BOINC projects,
> which crunch large numbers of numbers but don't take up a lot of space.
What is the best line for /etc/fstab ? The only example I have is :
'tmpfs /tmp tmpfs defaults,noatime,mode=1777 0 0'
This doesn't seem to limit the size in any way.
--
========================,,============================================
SUPPORT ___________//___, Philip Webb
ELECTRIC /] [] [] [] [] []| Cities Centre, University of Toronto
TRANSIT `-O----------O---' purslowatchassdotutorontodotca
