Google location question

Hank Nussbacher hank at interall.co.il
Wed Nov 24 06:36:23 UTC 2021


On 23/11/2021 22:31, Chuck Church wrote:

Old issue.  Everyone encounters this at some point:
https://www.iucc.ac.il/en/blog/2021-05-google-geo-location/

You can try reporting it to Google:
https://support.google.com/websearch/workflow/9308722?hl=en
and wait a month or so to see if the issue gets resolved.

Regards,
Hank

> NANOG,
> 
>                  Sorry if this is the wrong forum, but I figured we 
> could all use a distraction from IPV4 expansion for a bit.  We’re facing 
> a problem with corporate use of browsers and google location services.  
> Our users across North and South America seem to be getting location 
> info for Rio de Janeiro, BR as of the last few days.  All of our users 
> when browsing hit general internet via a NAT pool for IPs out of 
> Atlanta, GA.  That hasn’t changed in over a year.  The IP space itself 
> and the upstream routers PTR haven’t changed for longer than that.
> 
>                  For a normal laptop without a GPS chip and using 
> default settings, it would seem that something else must be informing 
> google of location.  We’ve had a few theories.  I discovered our Cisco 
> wireless in RIO office is broadcasting SSID.  The same SSID name we use 
> in all offices.  But I believe that google uses BSSID which is MAC 
> based.  And would be different between sites, as each have their own 
> wireless controller, and certainly different APs.  At least that’s what 
> I believe to be true, that google didn’t associate our standard SSID 
> name with a physical location.
> 
>                  I’m curious if anyone has ever dealt with this before.  
> I don’t want to blame our workstation folks for something that might be 
> a geolocation issue, either based on wi-fi being detected, or a messed 
> up geolocation for our IP space.  Is there any place in Google-land 
> where you can see exactly why it thinks you’re in a given location?  I 
> did see some articles on API usage to view values, but nothing that 
> shows you the process when it takes all variables into effect.
> 
> Thanks,
> 
> Chuck
> 



More information about the NANOG mailing list