From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly) Subject: Re: Wierd behavior with 0.98. Date: 2 Oct 1992 12:55:54 +0800
hodgen@infko.uni-koblenz.de wrote:
: Greets Linuxers!
:
: Last night I grabbed the 0.98 sources, added the fast scsi patches posted here
: recently and compiled (Jump table 4.1). Rebooted, start up X and go to xterm
: 1. Changed into /usr/src/pbmplus/libtiff, uncompress everything and edit
: Makefile. So far so good, now comes the fun bit. I type make and gcc starts
: compiling tif_fax3.c, lots of disk activity to start with and it trails off
^^^^^^^^ If I remember right, this is from xview ?? or something
like that??
The solution to your problem is to add about 45 megs of swap space...
GCC eats memory like nothing on earth, and 8 megs is NOT enough.
: slowly. I wait and wait, then I think, lets check the process table etc. Haha
: no cursor, no nothing! SOMETHING is eating ALL the CPU. I wait till the disk
When linux runs out of memory, and it has no swap space, it starts throwing
away the program so it can fit the data in. Eventally there is NOTHING that
it can throw away to bring some of the program back in so it locks.
Add some space space. 4 - 8 megs is good for normal usage.
: acitvity is nil then reboot and try again. Exactly the same thing happens. This
: time (1:20am) I think, maybe it'll come back by morning. Wrong! This morning
: (8:10) it was still locked up. Tonight I'll do it again and see if SHIFT-Scroll
: and CNTL-Scroll work (I forgot them last night - too tired). Also, after booting
: I ran fsck -v /dev/sdc3 (my /usr) and something had written about 15MB worth of
: Zone XXX marked as used but not used" type messages. Any ideas? My hardware
This was most likely the result of your reboot. Just because there is no
disk activity doesn't mean that disk it up to date. The buffers are
only flushed every 30 seconds.
:
: Wayne Hodgen | hodgen@infko.uni-koblenz.de | Opinions (c) Me 1991 | Intel
Michael