calcru: negative time (Interrupt Storm on cpu0: timer)

[J. Buck Caldwell]

Only time I've ever seen this problem is running on a VMWARE virtual 
server. Is that the case here? If so, try:

sysctl kern.timecounter.hardware=TSC

That did the trick for me.

Frank Steinborn wrote:
> Hello,
>
> After reboot yesterday, my system suddenly started to flood syslog with
> following message:
>
> calcru: runtime went backwards from 1966191374 usec to 1966189453 usec
> for pid 29988 (screen)
>
> It's not only screen, i saw sh and su also, but in the last hour only
> screen. A vmstat -i and the same 5 seconds later:
>
> interrupt                          total       rate
> irq1: atkbd0                           1          0
> irq13: npx0                            1          0
> irq14: ata0                       410366          5
> irq17: em0                      20882069        263
> cpu0: timer                    158483081       2000
> Total                          179775518       2268
>
> interrupt                          total       rate
> irq1: atkbd0                           1          0
> irq13: npx0                            1          0
> irq14: ata0                       410430          5
> irq17: em0                      20885602        263
> cpu0: timer                    158505514       2000
> Total                          179801548       2268
>
> Is this an interrupt storm? I know that there is a FAQ entry about the
> negative time issue, but it seems this is more a bad workaround I 
> guess. The system was running without problems for 3 months or so and
> as I said the problem suddenly appeared yesterday.
>
> Any hints on whats wrong here?
>
> Thanks in advance,
> Frank
>
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org"
>
>
>