My PC reports “An error occurred while Windows was synchronizing with time.windows.com.” or “…time.nist.gov”. How do I fix this?
The Help and Support Center suggested by the Date and Time Properties dialog says if it doesn’t work, it’s probably because NTP is blocked by my router. I go through a Linksys BEFSX41 router whose manual says it supports NTP. That’s the only reference to NTP in the manual. The config page for the router does not list NTP as one of the services that can be controlled and does not mention it in any other way.
I also go through a “Motorola Surfboard” cable modem to my cable ISP. I haven’t found any reference to NTP or network time protocol on those config pages either.
Is there some way to find out where in the system synchronization is failing?
Q.E.D., you got it. I added one of Dave Mill’s services at the University of Delaware, my alma mater as it happens. Dave created NTP in the first place.
Fiddlesticks, it was dumb of me not to say, but I use the hardware firewall in my router and have the Windows firewall turned off and no 3rd party ones installed. I’ve had problems with the Windows firewall preventing many things from working.
Zipper, you might be right about the timeout. What QED suggested results in my adding another server and selecting it.
Now I want to know why my freshly synchronized PC clock runs 14 seconds fast relative to my handheld GPS unit. One issue would be if the GPS didn’t have a recent version of DT1, the offset between UTC and GPS Time. This is downloaded from the GPS satellite every 12 minutes and I left the unit on outdoors for about an hour before using the value, so it should be fresh (I don’t know how to tell if it is). Since this is the last day of the month, there wouldn’t have been a leapsecond since the end of September (if there was one then), and I have used this device earlier in the month so it should still have a good value for DT1.
Never mind. I went to http://nist.time.gov/timezone.cgi?Eastern/d/-5/java which runs a web page showing a digital clock from their time service independent of my PC clock, and my PC clock tracks it within about half a second. So whatever is off by 14 seconds - no, wait, it’s actually closer to 15 - well, it’s the GPS. I think the missing DT1 value is about this big. maybe 18 seconds? but I don’t remember which way.
buckgully, i don’t think it’s that simple. I checked and found various references saying that GPS is 14 seconds ahead of UTC, whereas my computer is 14 seconds ahead of my GPS unit, which is the wrong direction. Also, I know that my handheld GPS unit is using GPS time internally, but it says it is displaying UTC. The manual says little about the distinction, but the manual for another GPS unit from that manufacturer says the unit provides UTC based on the number of leap seconds difference between GPS and UTC as broadcast by the GPS satellites, and it can also tell you what the difference currently is. I don’t have that unit set up to run at the moment but I may set it up and run it to see what it says.
Here’s an engineering comment from Garmin, the company who made both my devices, and posted at gpsinformation.net:
Provided the unit has collected current leap second count from the navigation message, (current leap second difference from GPS time is only broadcast once in a 12.5 minute Nav. message), or current leap second has not changed since the last time the unit collected this variable, the time displayed on the front of the unit should be accurate to within 1 second of UTC.
No, there’s something else definitely going wrong. When I checked again, my PC was 56 seconds fast relative to my handheld GPS, but the GPS agrees with my corporate laptop. When I resynch the desktop PC it goes back to being 14 seconds ahead of the GPS (NOT 14 seconds behind it). And the NIST website that displays the current time is 14 seconds ahead of the GPS (NOT behind) and the laptop. I am pretty much baffled as to what I can trust.
As my friend who worked in Time Services at the Naval Observatory (and who used to fly the atomic clocks around the world for various reasons) used to say, a man with one clock knows what time it is. The man who has two clocks has no idea.
>a man with one clock knows what time it is. The man who has two clocks has no idea.
Commasense, you owe it to yourself to check out Tom Van Baak at leapsecond.com.
Anyway, I farted around with 3 GPS units, 2 PCs, 3 autorefreshing web pages run by government time services, and the voice telephone version of WWV at the USNO, and got everything to agree within a second, except 1) the corporate laptop seems to march to its own drummer and has greyed-out internet time synchronization, and 2) my desktop’s built in clock is just a hair better than me saying “One one thousand, two one thousand, three one thousand…”