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.

Hi Christiaan

If you need to keep your time-series data on the new instance you will need to use the full backups on the SQL and NoSQL DBS.

If time-series data is not required you can use the JSON export and just move over what you require.

Hi @Robmalone

You should also look at the document for <ma-point-statistics> it returns a statsObject which you can then use statsObject.last, statsObject.max etc. this would be a lot easier to use than <ma-point-values> you for purpose.

Thanks guys for checking. I wasn't sure if the new core upgrade was checking the database name and version.

Cheers

Brian

@iperry you have not missed anything, 1 attempt is made, there is no setting for the number of attempts.

Nicely done man!
At least you know why now. Glad I could be a sounding board!

Fox

@aniluoh said in QModMaster Simulator for Mac OS:

Simulator
QModMaster
ModBus

Looking for the mac supported simulator.

Basically you cannot represent 30.03 exactly in binary.

The publisher is sending the raw value of the data point, which is represented by a Java double as the value you are seeing sent out 30.029...

When using a TextRenderer (which only happens for display purposes) you can set it to round the value via the pattern on it. That is why in the UI it looks correct but it is actually stored as 30.029....

(11110.00000111101011100001)₂ = (1 × 2⁴) + (1 × 2³) + (1 × 2²) + (1 × 2¹) + (0 × 2⁰) + (0 × 2⁻¹) + (0 × 2⁻²) + (0 × 2⁻³) + (0 × 2⁻⁴) + (0 × 2⁻⁵) + (1 × 2⁻⁶) + (1 × 2⁻⁷) + (1 × 2⁻⁸) + (1 × 2⁻⁹) + (0 × 2⁻¹⁰) + (1 × 2⁻¹¹) + (0 × 2⁻¹²) + (1 × 2⁻¹³) + (1 × 2⁻¹⁴) + (1 × 2⁻¹⁵) + (0 × 2⁻¹⁶) + (0 × 2⁻¹⁷) + (0 × 2⁻¹⁸) + (0 × 2⁻¹⁹) + (1 × 2⁻²⁰) = (30.02999973297119140625)₁₀

@terrypacker said in Datafile Source - Datetime is in UTC:

.withZone(DateTimeZone.UTC);

Terry Packer for the win! Thanks for the amazing help.

@tungthanh500 said in Device disconnect and data logging:

From 3PM to 9PM, the server is disconnected so I expect the value should be 0.
Is there are any setting so I can make sure the logging value is 0 if server/power meter disconnected, therefore I or anyone can realize the abnormal and check the server/meter?

Setting a point that is unhealthy to zero. would have a lot of unwanted consequences and create a lot of noise in your data when doing analytics on it... The point does indicate that it is unhealthy, just not on the chart. The live value will have an exclamation mark next to it.

Also even if you do not log the point when it is unhealthy, your chart will still have the same shape... IF you want to be get notified when the point goes unhealthy you can set a "has not updated" event detector on the point and link the detector to an email event handler.

Check your ma.logs, if this thing is internet facing someone may have been hammering it causing the login to drop out in relation to max non-authenticated attempts being tripped..

Fox

The latest DNP3 module should solve your problem. Let us know if you have any problems with it.

https://store.infiniteautomation.com/module/dnp3

This wholly depends on if you're using the SQL database to store values or the NoSQL database because it's on mango unit or a cloud installed setup.

I'd wager you're using the NoSQL system which means you're gonna have to export, open with notepad, copy and paste the data to maintain the correct date-time format (excel likes to screw with it) then filter by the value column.
With all of your filtered values, append the delete value in the final column, save, and reimport values to delete all unwanted values.
Also note you can change your datapoint properties to filter and ignore extreme/unwanted values so they won't go into your system.

Hope this is of some help

Fox

@meverett without knowing exactly what the problem is it will be hard for us to diagnose/fix the bug. Can you provide a pcap file from WireShark for the communication for a few polls? It is either the read request for the data points OR the connection request to initialize the connection.

Once I have some time I will take a look. I will admit it has been some time since I've decomposed any ABeIP packets but hopefully it will come back to me quickly!

Thank you @terrypacker. I had a global script using $EXEC
Once I removed that the script would execute.

Thanks for the response--

That seems unfortunate.. I know mango has a USER-CREATE-REPORTS type role for this kind of thing for report creation-- I don't think it's unreasonable to have a similar permission role for my users to be able to configure their own alarms.

I suppose for now, that's where we are, but that means I need to hand out lots of superadmin to people, which is definitely not ideal.

Heya:

I ran into similar issues with memory usage-- Notably, whenever one of our users ran one of their Big Excel Reports, the Available Memory would go to Zero, and the wheels would come off of Mango. We set our JAVAOPTS to be 10G, and those problems seem to go away.

Mango uses Monster Amounts of Memory under certain conditions-- I'd say that you 1600Meg is too small. I've got site machines running with 2 gigs of RAM, and we use the JAVAOPTS small for that. Those sites only use the Modbus datasource, a handful of metapoints, and the persistent publisher. Even that seems to use all the available memory at times, and once Java takes RAM it never give sit back..

Right now, our main machine is set up with 16 gigs of RAM, with only 174 megs free. The rest of that is JAVA, which is using 14.785 GB of RAM according to TOP.

This peaks whenever a report is ran, but then Java never gives RAM back to the OS after it's taken. Right now, our JVM free memory is about 5 gigs, but that will go to < 2 gigs if one of these big reports gets ran.

Cheers,
-Greg

We do it via json-file /administration/JSON store.

If you're using a single value, try
{{pt.value/3600}}
or pt.value/3600
depends on your markup

Fox

Hi @shoffend from the logs that you sent to us it looks like all you need to do it delete the Brewersdashboard module and restart mango.
It is located in /opt/mango/web/modules

Thanks Craig,