It seems that the kerneld loads in the modules when being used, "free" will report the total memory as that available after the kernel has been taken away - or at least thats what I have been told ! Pete ---------- Forwarded Message ---------- Subject: Re: [SLE] Complete system lock up Date: Sat, 4 Mar 2000 20:44:56 +0000 From: Pete Atkinson Its getting stranger ..... if I don't have kwintv running, I have 62880 ram, if I start Kwintv it drops to 59680, I thought 'free' would have reported ram regardless of what is running - anybody else like to try with this ? I suppose at the end of the day, I should get into the habit of killing seti before I try and run anything - not exactly ideal though..... Pete...... aying that, i have now got kmail, kwintv, seti and a kvt window open and it still shows 59680 with 1780 free - no sign of anything freezing ;-)

Two thoughts: 1) I have 64MB of RAM and my free shows: total used free shared buffers cached Mem: 62640 61068 1572 59580 924 24052 -/+ buffers/cache: 36092 26548 Swap: 82652 528 82124

yours shows only 59690. Where is the rest?

2) If you are able to do free and top, why don't you 'kill -9 2162' to stop seti?

JLK

On Sat, 04 Mar 2000, Jerry L Kreps wrote:

Pete Atkinson wrote:

...

Thanks to all that have replied..

I have tried to telnet in, but unfortunately even though seti etc is still running, the "system" is dead - I can't even ping my machine..

I have also offloaded squid and wwwrun which seemed to be hogging a fair amount of ram - I have "niced" seti as this was quite intensive but it still does it (occassionally)

So..

from a ram point of view, this is free

total used free shared buffers cached Mem: 59690 56512 3168 33540 1980 19112 -/+ buffers/cache 35420 24260 Swap: 128516 3028 125488

Considering that Seti and the Tv are currently running, the swap use seems low ?

This is extracts from top

PID PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM COMMAND 2162 16 11 13628 13M 600 R N 0 98.6 21.9 seti 2043 0 0 12112 11M 1224 S 0 0.3 20.2 X 2156 0 0 8252 8252 3412 S 0 0 13.8 kwintv

finally, my fstab

/dev/hdb4 swap swap defaults 0 0 (128M) /dev/hdb3 / ext2 defaults 1 1 (6GB) 66%cap /dev/hdb2 /boot ext2 defaults 1 2 (8MB) 26% /dev/hda1 /dos1 vfat defaults 0 0 (850M) 79% /dev/hdb1 /dos1 vfat defaults 0 0 (1.6GB) 38%

I sippose with the benefit of hindsight that my disk set up is all wrong - especially now that windross is effectively redundant - it seemed to evolve that way I am using a 400Mhz Celeron - 64M Ram on a 66Mhz LX board

any ideas -

Pete

On Wed, 01 Mar 2000, Iwan Mouwen wrote: > Ryan Yagatich wrote:

...

...

I've had somewhat of the same problem before. But mine was because I ran killall5

it sounds like your ram is being over-used. Try killing some processes and see if that helps at all

ryan

Or, if you're on a network, try to telnet (rlogin, ssh, ...) into your machine, and see what's going on.

Iwan.

-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/Support/Doku/FAQ/