Hypertrend and system time

[bkinard@kinardcontrolsyst]

I recently upgraded from an earlier version of Lookout to 6.1. After updating, I noticed that the ActiveX hypertrend times did not match the system time. I deleted the entire database and also deleted all of the hypertrends from the project; then I recreated the hypertrend objects. This still did not fix the time. The computer is a standalone unit and does not use the time sync feature. The attached picture actually shows that the hypertrend time is about 1 hour ahead of the system time. Any suggestions? Ideas? Solutions?

Message Edited by bkinard@kinardcontrolsystems.com on 04-10-2008 02:27 PM

[Ryan.S]

What's the lookout time? You can find it on the bottom left corner in run mode.

The hypertrend gets data and time from database, so I guess the time in Lookout is also incorrect. This may not be the hypertrend's issue. 

[Bob_L]

I've just run into the same situation. Periodically the Lookout time jumps 1 hr 11 minutes ahead and somtimes 2 hrs 22 minutes. Within a short while the time reverts to the correct value.

This creates a number of problems. Firstly, data values are logged at a future point in time. Once the clock resets to the correct time, values are logged again which if displayed on a hypertrend ends up looking like 2 lines.

Alarms that occur during the time the clock is advanced end up being set AFTER they are reset.

W2000 SP4

Any resolution?

Bob

[L Henry koile]

I have same problem and found that my Bios time was getting corrupted-  I though that had it fixed ( by resetting the time in the bios) but not ???   I will still happen   Why I don't know yet!!!   I saw in one of the other threads to the same problem about the AMD " fix" .  I just downloaded it and will try it tomorrow

Lookout 6.1    dell Vostro 1000- dual core AMD 64   It really drives my hypertrends silly.  I have to fix this, as my client is not impressed

NI guys what have you got

Best Regards

L Henry Koile

[Ryan.S]

The AMD processor does have a problem which causes the Lookout time to jump. You may see "Lookout time is advanced by xxx" or "Lookout time is set back xxx". The AMD fix solved this problem.

I'm not sure if you problem is the same one. Do you also see the event about time jumping in Lookout?

[bkinard@kinardcontrolsyst]

Still having the same problem with the hypertrend times, but with a new twist. Lookout is logging properly with the correct times but it jumps back one hour in time when viewing the data live.

Example: When I press the "end of data" button, the data is logged properly and the time is correct. But after a few seconds, the live data trend rolls back an hour. I can then press the "end of data" button again and it shows the trend correctly.

Its like the trend believes the end of the data is one hour behind.

This is the trend when I hit the "end of data" button. Notice end time on the trend, the system time, database time.

This is a few seconds (maybe 10 seconds) after hitting the "end of data" button with no interaction. Again, notice the times.

[Ryan.S]

Try to add the trace to hypertrend one by one. For example, add Smith Circle first, see if it has the problem. And then add Knox Road... Maybe the problem is in one or several traces. 

Open the MAX(Measurement&Automation) from desktop. Historical Data->Citadel 5 Universe, right click and create Trace view. Add the three traces to veiw. I want to know if the Hypertrend in MAX has the same behaviour.

[bkinard@kinardcontrolsyst · Accepted solution]

I tried opening the traces in MAX on a different system that is exibiting the same behavior except the trend jumps back to 1903. Max did the same thing as Lookout has been doing. It almost has to be the data because I have several different trend charts in this process. Some are correct, some go back to 1903 and some are actually in the year 2192.

It turns out I had duplicated some traces and pointed the trends to the wrong trace. Once I corrected that problem, the times were corrected also.