Weird rtc interrupt rate

[Dan Nelson]

In the last episode (Jun 22), Jeremie Le Hen said:
> I have a Dell Latitude CPx laptop running FreeBSD 5.2-CURRENT.  I
> have been using FreeBSD on this laptop for a long time, and the
> problem I'm going to describe also happened while it was running
> 4.6-STABLE.
> 
> Sometimes, when wathing at my processes using top(1), all CPU states
> are 0.00.  Since ps(1) shows the same behaviour, this is not a bug in
> top(1).  But when this problem happens (this seems to be completly
> random), vmstat(8) shows a weird interrupt rate for rtc :
> 
> # z6po:root# vmstat -i | sed -n '1p; /rtc/p;'
> # interrupt                          total       rate
> # irq8: rtc                        7584528        115
> 
> or, two hours later, while the computer was 100% idle :
> 
> # z6po:tataz$ vmstat -i | sed -n '1p; /rtc/p;'
> # interrupt                          total       rate
> # irq8: rtc                        7584528        101
> 
> AFAIK, rtc interrupts rate should be something like 127 or 128.  What
> could make the rate being so unstable ?

It's not unstable; it just stopped completely and the average is slowly
falling.  I had this happen to one of my servers a few weeks ago as
well.  You can determine exactly when it stopped by counting
7584528/128 ~= 59254 seconds forward from the time your system booted. 
Mine died after about a week of uptime, and I didn't see anything
interesting in the logs.  A reboot fixed it.

-- 
	Dan Nelson
	dnelson at allantgroup.com