streams of set_rtc_mmss: can't update from

Hmm well the error message is in response to the CMOS clock being set. If you are running ntpd it will periodically try and sync the CMOS clock to that of the more accurate system clock. However something appears to be going wrong and your hardware clock is vastly out of sync with kernel's clock.

See http://www.ntp.org/ntpfaq/NTP-s-trbl-spec.htm#Q-LINUX-SET-RTC-MMSS for some details.

Dave:

What happens if (as root) you do
ntpdate ntp.ubuntu.com
hwclock -c

Do the messages go away?

* Sitsofe Wheeler (<email address hidden>) wrote:
> Hmm well the error message is in response to the CMOS clock being set.
> If you are running ntpd it will periodically try and sync the CMOS clock
> to that of the more accurate system clock. However something appears to
> be going wrong and your hardware clock is vastly out of sync with
> kernel's clock.
>
> See http://www.ntp.org/ntpfaq/NTP-s-trbl-spec.htm#Q-LINUX-SET-RTC-MMSS
> for some details.

Thanks - hmm; OK - my hwclock is about 4 hours out, which I guess
explains whats going on. (I'd actually noticed the hwclock was out
a few weeks ago and did a hwclock --systohc - it looks like its
drifting off really quickly - probably time for a new battery).

> Dave:
>
> What happens if (as root) you do
> ntpdate ntp.ubuntu.com
> hwclock -c
>
> Do the messages go away?

So the messages are coming and going anyway; I guess depending how far
things are out at any one point - I've just done the ntpdate
and hwclock (although -c doesn't seem to exist so I used --systohc)
and we'll see what happens.

So this is obviously not a bug then; but I'd say two things:

   1) I think its bad form to log something every few seconds;
that probably would cause major pain for someone trying to spin
a disc down (and probably slow things down).

   2) Shouldn't the system do something more noticeable for a user
to tell them their hardware clock is out? (Not a kernel issue,
but a 'Linux for Human's' issue)

Dave
--
 -----Open up your eyes, open up your mind, open up your code -------
/ Dr. David Alan Gilbert | Running GNU/Linux on Alpha,68K| Happy \
\ gro.gilbert @ treblig.org | MIPS,x86,ARM,SPARC,PPC & HPPA | In Hex /
 \ _________________________|_____ http://www.treblig.org |_______/

Setting back to unconfirmed after Dave's reply.

As for telling the user - I don't know how many users are technical enough to fix a hardware clock error even if it is just a case of swapping the battery (in some cases that can lead to the BIOS being reset etc.)

This bug has had no activity for a considerable period. This is a check to see if there is still interest in investigating this bug report.
It appears that it may have been related to a hardware fault.

Closing.

What about th bug? It's still there!!

May 9 10:58:27 www kernel: [77529.208283] set_rtc_mmss: can't update from 3 to 58
May 9 10:58:28 www kernel: [77530.208459] set_rtc_mmss: can't update from 3 to 58
May 9 10:58:29 www kernel: [77531.208282] set_rtc_mmss: can't update from 3 to 58
May 9 10:58:30 www kernel: [77532.208346] set_rtc_mmss: can't update from 3 to 58

Work around to the bug:

sudo hwclock --systohc