Re: `top` reports all CPU states as 0% on i386

From: Jason George <bsdlists_at_masterplan.org>
Date: Mon, 12 Jan 2004 01:32:44 GMT
>I think this problem is usually caused by one of the clock interrupts
>stopping. Checking the output of "vmstat -i" can confirm this - check
>for:
>
>clk irq0               1044322763        100
>rtc irq8               1352668148        129
>
>The rate for the first one should 100 (Or possibly HZ, if you have
>adjusted HZ). The rate of the second should be about 128, unless
>things have got confused.
>
>        David.

Yikes!  I think my RTC has left the building!

> vmstat -i
interrupt                          total       rate
irq1: atkbd0                           1          0
irq6: fdc0                             4          0
irq10: fxp0 xl0                    86393          1
irq11: atapci0                    255394          3
irq13: npx0                            1          0
irq0: clk                        8087632         99
Total                            8429425        103
>


Another point of interest is that about 8 months ago, I added 
kern.timecounter.hardware=i8254 to /etc/sysctl.conf because the system clock 
was going insane - it was counting off an hour of system time in the matter 
of a minute or so of wristwatch time.

--Jason
Received on Sun Jan 11 2004 - 16:32:49 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:37 UTC