Please Note This forum exists for community support for the Mango product family and the Radix IoT Platform. Although Radix IoT employees participate in this forum from time to time, there is no guarantee of a response to anything posted here, nor can Radix IoT, LLC guarantee the accuracy of any information expressed or conveyed. Specific project questions from customers with active support contracts are asked to send requests to support@radixiot.com.

@MattFox I double checked the menu item permissions and it was correct @phildunlap This was exactly it. I overlook the JSON every time. Thank you!

No problem, I'm glad you came right!

That makes more sense than it being a timeout. A timeout would have produced a ModbusTransportException with timeout as its message, not an operating system message about a file descriptor having issues.

@petermcs you can move the position of the text with offset properties

Hi luigi, Thanks for putting so much detail in! The only thing that's missing is the setup for the high limit detector you have on DP_i1. Your output point looks like a virtual point, so there is no excuse for it not accepting a set (unlike a Modbus point for instance, which could have the set request timeout). It is very weird that you have that virtual point set to an 'Alternate' change type. This means every poll period of the virtual data source it will flip from true to false or the other way around. Have you disabled polling on the virtual data source that owns DP_Ai1? I would look at your events to try to figure out what is going on if it isn't a misunderstanding of the change type. I would expect a message about the SetPointWorkItem failing if you see all the events raised / returned that you expect to. If on the other hand you don't see all the high limit events that you were expecting on DP_i1, then I'll wonder what version of Mango you are using, and I'll want to see the settings on the high limit detector.

Brought them out of root level and now everything is great. Have some buttons assigned to specific date ranges on the page itself. Now if there's a way to hide the top blue bar, it will be perfect.

Hi andresmorago, I don't have a great deal of experience with OPC devices, but some googling and reasoning suggests it's either an error with the server not having sufficient permissions (seems unlikely unless you installed the OPC server software as well, then possible) or it has some sort of authentication you need to get Mango to do or a client whitelist that you need to add it to.

@pierjsap Please do share your solution as I can see this logic being very useful. I think my method would be to duplicate your event sent points and logic in the meta data point script. Then use the if statement that your proposed to set the meta point. Keep the event detectors so that your event page will still list the events.

Hi Wingnut2.0, Yes! Unfortunately, the modules' i18n.properties files in their classes directory is not affected by the overrides directory currently. This means that currently you need to modify that file in place, then back it up for when updates come along. To change the alarm names, you can copy Mango/classes/i18n.properties into Mango/overrides/properties/ and modify it there. After you've renamed the alarm levels as desired, you can restart Mango to see the changes. It's also not an option to simply put the updated keys into the one i18n.properties file, either. You have to modify the module's i18n.properties file.

https://github.com/infiniteautomation/ma-core-public/issues/1236

thanks @phildunlap ;)

Hi Wingnut2.0, A quite large enterprise server I've seen, the one described here, has a 45GB ext specified heap, 40 cores at 2.6 GHz. About 550000 total points across 30 or so persistent data sources, 300000 values/second, throttle incoming persistent threshhold set to 37500000. Also a very large NoSQL minimum batch size (so that it always hits the minimum small batch wait time, which is 60000ms on that system). This system has been cruising along for quite a while. The settings left us a fair amount of room to run large, complex reports or zip around the UI.

Yes

That makes sense, thank you!

@phildunlap Thank very much, The problem was resolved

Glad to hear it :D As I'm sure you saw, you can probably drop all the URL parameters for hiding things now.

@rectifier I agree, good idea.

Thanks for sharing the resolution!

If Haystack is available in that Niagara version, you would create a Haystack data source in Mango which would behave as a client and poll the Niagara server. This solution would probably be easier, go ahead and try! All you need is the URL to your Haystack server and your authentication information to create the data source. If using a scripting data source, you would have the full flexibility of the REST API that was the subject of the post. If it's serving a REST API and that's what you're querying, then you're the client and its the server again. There are other options. Both the BACnet and Modbus publishers in Mango accept writes over those protocols. If you find it easier to output a CSV once a day, there is a Data File data source that can import something you copy to it or you can define an arbitrary poll class and have Mango copy the file from the remote server. So, either client-server relationship is possible, but both the scripting and Haystack options would definitely be clients to their respective APIs.