Linux 2.0.30 (harik@chaos) (gcc ) # [chaos]
Memory: Total Used Free Shared Buffers Cached
Mem: 61736 19688 42048 11168 348 4176
Swap: 130748 48852 81896
Bootup: Sun May 11 04:05:51 1997 Load average: 1.33 1.97 1.73 2/113 8921
user : 9:39:37.31 5.2% page in : 9501591 disk 1: 1426137r 989600w
nice : 0:05:56.13 0.1% page out: 4776392 disk 2: 1461973r 828671w
system: 4:31:25.71 2.4% swap in : 168913 disk 3: 98459r 36021w
idle : 7d 3:41:45.74 92.3% swap out: 92877
uptime: 7d 17:58:44.86 context : 171120856
irq 0: 66952489 timer irq 8: 0 rtc+
irq 1: 1002587 keyboard irq 9: 4537232 BusLogic BT-956C+
irq 2: 0 cascade [4] irq 10: 117838345 SMC Ultra
irq 3: 45272 irq 11: 5345 fdomain+
irq 4: 4143559 serial+ irq 12: 0
irq 5: 0 irq 13: 1 math error
irq 6: 2 irq 14: 303689 ide0+
irq 7: 1941423 sound blaster irq 15: 0
Looks like the kernel is aggressivly paging programs out to make room for
buffer cache... to the point that I can't compile (gcc: exausted virtual memory)dmesg only has complaints about being unable to allocate a DMA buffer for
sound. I see the latest stable version is still 2.0.30... is there a pre-31
patch out I could test?
BTW: I can also reliably bring a completely unloaded 2.0.30 box to it's knees
by doing badblocks -w on a scsi device (obviously, one with nothing important
on it.) I have yet to be able to check the disk, since it runs out of
memory (with plenty of swap free) before it finishes checking...
I'm getting tired of having my xterms swapping out when I have 48 megs of
free ram... back to .29 for now.
--Dan