I have an issues with PI OPCDA interface version 18.104.22.168 which in configure as HOT Failover between 2 server.
After failover, some tags got value update but most of it the value not updated. Especially, Digital Tags.
Anyone can explain why?
can you describe the problem a bit more in detail? There are some open issues with this Interface version (for example we run into this: 121448 - Advise tags stop updating when excmax is surpassed )
If it is urgend please contact TechSupport for a quick solution.
The interface have been configured as Redundant. There is 2 server that perform HOT Failover between this 2 interface. If one interface loss the connection from Data Source, another one will take over. As a HOT Failover both are collecting data but the active one will send data to PI.
The issues is, when the failover happen due to loss connection to data source, it will write COMM FAILED temporary before it change to the other interface. The issues is, when the interface is failover only some of the tags will update with the value, other still have a COMM FAILED value. This only happen when failover.
But if the interface is independant, if loss the connection to Data Source, it will write COMM FAILED in value. But after the connection back to normal, its automatically update the value from teh Data Source.
It sounds like the issue is which slowly updating tags (Advise tags, long scan classes?) that are not getting a new value to write following the COMM FAILED value. Is that correct?
If so, we can use the OPC Interface's 'Data Handling' parameter "Update Snapshot (/US)".
PI Interface for OPC DA - Update Snapshot:
If the current snapshot is a system digital state (such as I/O timeout, Shutdown, and so forth) and the interface reads in a new value that is older than the snapshot, the interface sends the new value one second after the snapshot time stamp of the system digital state. This check is omitted if the current snapshot is a good value. This is useful for setpoints that rarely change. (/us).
Yes. I also thought the same thing. I turn on the Snapshot Update. But there is no update. quite a long time. im not sure why.
I test it again. Now with the exception max 600. Some of data got update after 5-10 minute after failover. And its take 20 min to update all the tags if the tags is same as previous value.
We could see if tweaking the ExMax value, or increasing scan frequencies (if polled) would play a role here.
Another option we'd have would be to "Send only GOOD quality data (/SG)", which can be configured on the OPCInt tab of ICU. This should tell the OPC Interface not to send any Questionable or Bad quality data to its PI tags, which would include something like "Comm Failed"
If we are to use this option, our archive data would be uninterrupted based on communication issues, but I'd strongly recommend configuring the DeviceStatus and other Health Tag data for the interface so that we can still identify if the interface is experiencing issues, etc. (since it wouldn't be blatantly apparently from a general OPC PI tag)
I try that also, but my customer still want to see if the communication between data source and and PI Node.
The one am not sure is, why its take so long to update when do failover. It is working fine when the same scenario happen to standalone server.
Do you have any explanation?
Retrieving data ...