Well this didn't make it fail... suspenders:~# ping -f localhost PING localhost (127.0.0.1): 56 data bytes . --- localhost ping statistics --- 8626135 packets transmitted, 8626134 packets received, 0% packet loss round-trip min/avg/max = 0.0/0.0/110.2 ms
neither did this suspenders:~# time cat /dev/ide/host0/bus0/target0/lun0/disc > /dev/null 32.630u 96.580s 20:31.01 10.4% 0+0k 0+0io 91pf+0w suspenders:~# time cat /dev/ide/host0/bus1/target0/lun0/disc > /dev/null 68.610u 214.530s 33:22.55 14.1% 0+0k 0+0io 92pf+0w I'm looking at the 512 Mb dimm lying on the table and guessing that was the one that was faulty. On Fri, 2002-10-25 at 09:25, G. M. Bodnar wrote: > Tim Wright is on permanent record as saying: > :On Fri, 25 Oct 2002, C Falconer wrote: > : > :> Whats a good way to simulate load on a server? I've flood-pinged it for > :> 10 minutes, and the staff are complaining about network response to the > :> other server! :) > : > :you could compile the kernel, or do something like this under bash: > > [snip] > > :of course, that'll load up your CPU not your memory. > > I've found editing large audio files to be a bit of a memory hog. > Something on the order of 200MB dumped into memory and flipped around. > > Alternatively, for disc and CPU access, I had pretty effective loading > last night with a sin approximation test that I was running over 1024*8192 > iterations and dumping the results to file... > > Greg > --- -
