I would like to be able to configure two identical instances of the e-terrahabitat connector on different computers, only one of which collects data from habitat at any given time. In the event that the primary instance or the machine that hosts it fail, the backup instance must begin collecting and sending data with minimal loss.

Enable UFL Connector to interpret two consecutive commas in a csv file to mean that the data field is empty. The UFL connector should interpret this empty field as a zero.

Connectors need adapt to many different types of networks, in both configuration and quality (latency, etc.) of the data transfer. One of the features that interfaces had, but connectors still lack, is the ability to batch the values sent to PI and spread out when they are sent. These are the Send Rate settings for the buffer subsystem, they are requested for connectors as well.

What we would like is a third option with UPDATE and INSERT, which prevents a new value for the same timestamp from being processed. Such as a NO REPLACE mode.

The EtherNet/IP protocol does not specify timestamps when data is transmitted, the connector currently assigns the timestamps base on when the data is being processed.

This leads to the following issue, if the network/computer/etc gets busy, we will have several values be processed at the same time. Thus, many values with the same timestamps.

A possible work around is that the connector specifies time frequency, for example values come 10ms apart.
If the connector read this frequency, the connector could possibly assign better timestamps for values that are processed at the same time.

Currently, the connector will create 124 or 248 tags per data source, regardless of how many meaningful data points are actually in the data set.
This greatly limits the use of the connector to any customer that is not an EA.

Option to discard Cygnet data with status unreliable so that only good data is sent to PI.

Add a Switch to the OPC UA Connector that we can choose if we want the AF Attributes to be created using "Browsable Name" or "Display Name". Using the Display Name would make it much easier for us to locate the Attribute and refer to its OPC Tag as we can see the same "Display Name" in both PI System Explorer and OPC Tools.

For the PI Connector for EtherNet IP you can have up to 100 data sources. Typing in all of them is tedious. It would be nice if we could copy/paste information from Excel OR have a structured import/export file to the data source list.

After latest patches Windows 2012R2, IE access to UFL Connector 1.3.0.106 admin pages is no longer functional.

As some IT departments do not support different versions of browser on servers, or support opening up ports for admin access, this limits ability to administer the UFL.

Please support the default server browser for the supported operating systems.

Develop a connector for Prometheus \ InfluxDB

Develop a connector for Splunk.

When going in the PI Connector Admin page in the data sources page, we cannot start/stop individual data sources listed when there is more than one. To stop one you need to stop all the data sources listed.

Currently only hot failover is featured for the PI Connector for OPC UA but this creates a lot of strain on the data sources. We need to be able to select warm/cold failover mechanism.

Is it possible to get it soon? it's not an option to set up a host for each system connector. it might be needed to have different data sources to define independent data streams for different use cases (e.g. send Data of AF Branch "A" to global Destination AF Database "A" and Data of Branch "B" to global Destination AF Database "B").

All in all a wrong configuration should be undone easily and not result in a complete re-installation of the system. If this is possible by using powershell scripts (an example is installed for all connections - RemoveAllConnectionConfigurationsRelay.ps1 - maybe it might be a base for a script to remove single registrations).