Noah
Robert Thoerncrantz wrote:
>
> Sometimes I find that load average rises whithout top showing
> any CPU eating processes. The system stays perfectly
> responsive as ever, just as if nothing special was going on.
>
> Usually it is stuck at one, and starting top to investigate
> gives a load average of about two.
>
> If I start the rc5-client, top shows it at >95% as it should,
> and load average slowly gets to almost 3. Then it drops to the
> more normal ~1.0 again and everything returns to normal.
>
> I'm trying to reproduce it on demand, but have not had any
> luck yet. Am I the only one that experiences this problem with
> load average?
>
> The kernel is a vanilla 2.1.130 (UP), but it happened in
> 2.1.129 (and, I believe, 2.1.128) too, then I thought it was
> just some weird effect from the UP-flu.
>
> Linux version 2.1.130 (rtz@pirx) (gcc version 2.7.2.3) #1 Fri Nov \
> 27 14:34:14 CET 1998
>
> 12:53pm up 2 days, 15:30, 3 users, load average: 2.19, 1.69, 1.28
> 49 processes: 47 sleeping, 2 running, 0 zombie, 0 stopped
> CPU states: 5.7% user, 2.8% system, 0.0% nice, 91.7% idle
> Mem: 30448K av, 29648K used, 800K free, 12040K shrd, 3216K buff
> Swap: 66524K av, 16836K used, 49688K free 11404K cached
>
> PID USER PRI NI PAGEIN SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
> 8471 rtz 17 0 233 748 748 568 R 2.6 2.4 0:08 top
> 251 root 19 0 19552 6776 5068 1548 R 2.4 16.6 76:52 XF86_S3
> 32753 rtz 8 0 6749 7664 2740 988 S 1.3 8.9 8:59 netscape
> 31889 rtz 7 0 1792 1156 812 432 S 1.1 2.6 9:43 wish
> 2111 root 9 0 6753 880 620 496 S 0.9 2.0 0:36 xterm
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/