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

Radix IoT Website Mango 3 Documentation Website Mango 4 Documentation Website

How to change the http sending value parameter

  • It is just a missing translations key. it was missing from the base translation file even, which is why you saw that instead of English.

    Thanks for bringing this to our attention!

  • please answer to the issue about the license that I sent you to support.
    Thank you

  • I have answered your email.

  • Now within MANGO v3 UI, once in awhile I see the following messages in the bottom screen:

    connectivity lost... ok
    (after 2 sec)
    connectivity to MANGO API as been restored

    My mangoApi module is 3.1.1

    Once mangoNoSqlDatabase is depending on it, could this be in any way related to the problem that we have? (please see bellow reminder text). Is there any log where I can see more info? I search in the ma-log and found nothing.

    let me remind you: when more than two simultaneous BACnet devices have it value switched only aleatory two reach's the http publisher. This is the only issue that holding us down to close the project... :-(

    Thank you,

  • Hi Jose,

    I probably can't solve the simultaneous alarms issue through the forum, since I don't really have a clear idea of what's going on and would have to investigate. You could contact us about direct support on that issue.

    You can probably get more information about why it's displaying that connectivity message if you watch the browser developer tools' network tab. My guess is some requests are timing out or erroring. You could also consider enabling debug logging by switching the log4j2.xml files in Mango/classes/ and reloading your logging settings, either in the Log4J Reset section of /system_settings.shtm or by restarting Mango.

    I would not expect it to be related to the dependency of the NoSQL database on the API. It just provided a new set of endpoints for saving point-time-jsonObject information, shouldn't affect existing things as I'm not aware of any built in features using that database yet.

  • I will come back to the simultaneous alarm issue after I collect some more info.

    meanwhile, after upgrade to MANGO v 3...

    A) In Home I can not see the number of datasources, datapoints, etc. in the up right corner of the square. Also, Active Sessions, Event Detectors and Mailing Lists are not clickable.

    B) !!systemSettings.lastUpgrade!! in Administration > System settings > System information, with 1500555481 as date.

    and this one, last two from ma.log?

    WARN  2017-07-24T14:50:10,495 (com.serotonin.m2m2.util.timeout.TaskRejectionHandler.rejectedTask:77) - Rejected task: User event cache cleaner because Task Queue Full 
    ERROR 2017-07-24T14:50:10,497 (com.infiniteautomation.nosql.MangoNoSqlBatchWriteBehindManager$StatusProvider.scheduleTimeout:726) - 7 BWB Task Failures, first is: Task Queue Full

    thank you very much,

  • A) The numbers will only appear if you have a configured internal data source. You can reinstall the internal data source and it should create it for you.

    B) This has been fixed.

    I wouldn't expect either of those log messages to cause you problems, but you can probably make them go away by changing...

    runtime.realTimeTimer.defaultTaskQueueSize=0 to runtime.realTimeTimer.defaultTaskQueueSize=2

    in your file and restarting Mango.

  • Hi Phil,

    I send you a message to support.

    Thank you,

  • I sent you a reply.

  • It was discovered that the HTTP publisher was sending all the point values, but it was putting multiple values into single GETs or POSTs, which the receiving end was not able to handle. Resolution was found by introducing an intercepting layer to send them on to the other program one at a time, as there is not a setting on the HTTP publisher to limit values per message (it's limited by URL length, post size, etc).

    Another option that was presented was using the point links to set the HTTP desired directly out a TCP/IP data point rather than tying the point links to a virtual point and having the HTTP publisher do the transport. This likely would have worked, but the existing solution was touched up before this was attempted.