Hi,
I have never seen a problem with kepware when requesting a rate that is different than the scan time of the tags defined there, so there seems to be something strange going on there. I suspect that the missed records can only be coming from either the quality going bad, in which case the trigger won’t be evaluated, or two events happening before the trigger reset is seen.
If possible, I would like you to send a XML export of the group, an export of the kepware project, and the wrapper.log file to support@inductiveautomation.com. This should give me a clear idea of how everything is set up and what is going on during normal operation.
Regards,
Hi Colby,
i will send you the Gateway backup and the Kepware file instead as i am off site at them moment.
Regards
Sascha
Where can i do this i don't see this setting anywhere in the ignition opc server setup on the gateway website.
I’m sorry, I might have been assuming something that I shouldn’t have- are you connecting to Kepware via COM or UA? If UA, you won’t have that option.
If you’re seeing messages about kepware being disconnected, that could be at the root of this, and the wrapper.log file should help us track it down.
Regards,
Hi Colby,
We are using Kepware OPC UA.
It says in the kepware log
Date Time Level Source Event
31/03/2011 23:07:46 Information UA Server UaServer_EndpointCallback: SecureChannel 1811610 renewed! [status=0x0]
This comes up every 30 minutes or so.
Will try and find time to get to site and get this wrapper log file.
The other thing i noticed is that:
In the transaction group when you configure the opc items you can not set the group to execute at any other rate then that of the update rate of the tags in kepware. If you don’t all opc tags in the transaction group become stale.
I can replicate this in a fresh virtual machine setup also.
Regards
Sascha
Hi Colby,
have forwarded the Wrapper log files from the system to support.
I have also taken a backup of the mysql database in case it is needed while i was there.
(120 mile rountrip wont be done everyday).
Hopefully we can get to the bottom of this soon.
Regards
Sascha
Thanks. I’ll reply via email.
Regards,
I sent you a fairly long email. Let me know if you don’t get it or can’t read it, sometimes our email system can be mean.
Regards,
Hi Colby,
got the mail and replied, thanks for the help.
Still an odd one out.
Regards
Sascha
Hi Everyone, Hi Colby,
First things first. Inductive Automation do a great job and are excellent in their support, Tanks Colby in particular to help me find this problem.
It was not ignition that was missing the record it is a scripting error of some sort.
The record that is stored by the transaction group uses the PLCDateTime which is made up of bytes in a Siemens PLC. The following script was used to convert the DateTime in the PLC to the correct format.
toDate(
2000+{[.]Year} + "-" +
{[.]Month} + "-" +
{[.]Day}+ " " +
{[.]Hour} + ":" +
{[.]Minute} + ":" +
{[.]Second})
It turns out that for some reason (havent figure out yet why) the script randomly returns:
Tue Nov 30 00:00:00 GMT 1999
Rather then the actual PLCDateTime.
This results in the illusion of a record being missed when one searches the database for PLCDateTime for a particular day.
Again thanks for the help guys in finding this nasty little error that has been haunting me.
Kind Regards
Sascha