SHR clock reset

[Ben Wong]

Bizarrely, my Freerunner also got the wrong time yesterday.  Is there
something special about October 1st?

I'm running SHR testing, but I booted into an old copy of SHR unstable
and it also go it wrong.  The first time the date was 1970 (the
epoch), the second time it thought it was 1933! (Negative time? I'm
curious how that happened.)

I'm hoping it was just that I was near a badly configured cell tower.
When I checked it again later, after I got home, the time and date had
reset itself correctly.  On the other hand, I have *never* seen a cell
in the US with the time misconfigured.  I'll test it again when I head
back to the same area next week.

--Ben

2009/10/1 Vasco Névoa <vasco.nevoa at sapo.pt>:
> I've started experiencing the same in the last few days.
> The Neo's clock had always been impeccably correct, until now.
> Now it was half an hour late and I set it by hand.
> It appears the clock is drifting, and otimed isn't keeping it in sync with
> the GSM.
>
> The only hints in frameworkd.log are:
>
> 2009.10.01 18:05:25.88 otimed               INFO     loaded timesources
> [<GPSTimeSource>, <NTPTimeSource checking 134.169.172.1 every 600 seconds>]
> 2009.10.01 18:05:25.124 otimed               INFO     loaded zonesources
> [<GSMZoneSource>]
> 2009.10.01 18:05:25.138 frameworkd.subsystem INFO     subsystem otimed took
> 1.53 seconds to startup
> 2009.10.01 18:06:27.163 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 18:06:27.357 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 18:16:39.457 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 18:16:49.835 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 19:18:44.462 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 19:19:20.392 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 19:19:48.627 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 19:20:40.194 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 19:20:58.312 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 20:49:58.117 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 20:53:14.628 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 22:18:28.542 otimed               INFO     GSM: multiple zones
> found
> 2009.10.01 22:48:41.491 otimed               INFO     GSM: multiple zones
> found
>
> Does this "GSM: multiple zones found" situation create a problem?
>
> I looked at the hwclock, but it seems it's not even in use / not related to
> the time shown to user:
>     root at om-gta02 ~ $ hwclock -l
>     Thu Oct  1 21:52:35 2009  0.000000 seconds
>     root at om-gta02 ~ $ date
>     Thu Oct  1 23:15:20 WEST 2009
>
> Any hints on what the problem is?
>
> Vikas Saurabh escreveu:
>
> Timezone from GSM is already implemented for ages ;)
>
>
> I have often got wrong timezones reported by cell (thereby resetting
> my phone's clock and making me reach late somewhere :(...but thats a
> different story)
>
> I was wondering if shr-settings can get something like:
> Use cell timezone:
> * yes
> * no
> * ask
>
> --Vikas
>
> _______________________________________________
> Openmoko community mailing list
> community at lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>
>
> _______________________________________________
> Openmoko community mailing list
> community at lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>
>