@jared-wiltshire said in Bulk Event Detection Creation:

when i try to use /rest/v2-full-event-detectors i get the below error

raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

however, event-dectors still works. Can you confirm the 'state' parameter should be set to the value that will trigger the event?

@Jared-Wiltshire do you have a data structure for creating an event detector? We're connected through the API and have been able to Get the existing even-detectors to look for a model as Swagger doesn't have a lot of information for POST /v2/event-detectors

Here is the response we get from /v2/event-detectors/data-point/{xid}

[
  {
    "xid": "ED_e7f4ceb7-baca-4842-b22c-897a4e2d87fe",
    "name": "PCS3 Has Faulted",
    "duration": 5,
    "durationType": "MINUTES",
    "alarmLevel": "CRITICAL",
    "supportedDataTypes": [
      "MULTISTATE"
    ],
    "handlers": [
      "EH_43fde3cf-1c12-483b-820c-9392c38f114e"
    ],
    "description": "PCS3 Has Faulted",
    "eventType": {
      "dataSourceId": 2,
      "dataPointId": 995,
      "pointEventDetectorId": 110,
      "duplicateHandling": "IGNORE",
      "rateLimited": false,
      "typeName": "DATA_POINT"
    },
    "sourceId": 995,
    "rtnApplicable": true,
    "detectorSourceType": "DATA_POINT",
    "validationMessages": [],
    "detectorType": "MULTISTATE_STATE",
    "id": 110
  }
]

What i dont see from the return of the GET is how we set the XID of the data point and the level. For example the above is set by a data point on value 8. Im not seeing in the return the XID of the datapoint that triggers this event detector or to look for value 8.

What are we missing?

Thanks for these thoughts. We're review the JSON idea and Phili's last script.

Our other thought was to use the real time values every x seconds through API and check if type = binary and name contains "fault" then do some task... this would be outside of Mango though so its not 100% desired but could be a scaleable work around.

We're deploying a new system and the data source has about 1,800 fault bits (200 in each subsystem and 9 subsystems).

There are binary status bits that indicate faults or alarms of a subsystem. Creating the data points has been relatively automated through the CSV uploader, however, we cannot seem to find a method to automate the creation of level based event detection on each of these bits.

Does anyone have suggestions or thoughts on how we could automate the creation of these event detectors?

@phildunlap we have not seen that, we'll give it a try. thanks

We have an application where this is a dashboard that provides a high level view of 5 pieces of identical equipment. We also have a detailed page of the equipment that provides more data (meters, faults, alarms). In Mango we have 1 detail page that is then populated based on tags.

RIght now you have to select the tag from a drop down list. What we're trying to figure out is how can we set the tag from the link.

For example.. if someone is looking at the summary dashboard that shows all 5 pieces of equipment then they click on Unit ID#2 to go to the detail page we want the detail page to know to set the tag for Unit ID#2.

Does anyone have ideas on how to do this?

@phildunlap thanks this worked great

I have a watch list serial chart in a dashboard. I do not want the From and To to be populated by datebar.To and datebar.From instead I want it to always be the trailing 7 days and still enable the user to zoom if desired.

So ideally it would be like:
to: datetime.datetime.now()
from: datetime.datetime.now() - timedelta(days=7)

Any thoughts on how to achieve this are much appreciated.

Thanks

This fixed it and i can now login. thanks @MattFox and @phildunlap

payLoad = {'username': userName, 'password': password}
session = requests.Session()
session.headers.update({'Cookie':'XSRF-TOKEN=dc1d54de-fab1-4405-8ccb-ea677aadf7ca','X-Xsrf-Token':'dc1d54de-fab1-4405-8ccb-ea677aadf7ca','Accept': 'application/json', 'Accept-Encoding': 'gzip,deflate,sdch'})
# session.headers["Content-Type"] = "application/json"
response = session.post('https://monitoring.[domain].com/rest/v2/login', json=payLoad)

Thanks @MattFox for the quick response. swagger is enabled and mango has been reset but now I'm getting a 403 error. I have confirmed the username and password are accurate. Any ideas?

payLoad = {'username': userName, 'password': password}
session = requests.Session()
response = session.post('https://monitoring.[domain].com/rest/v2/login', json=payLoad)

if response.status_code == 200:
    print("Success Logging In")
    print(response.status_code)
    print(response.text)
else:
    print("Error")
    print(response.status_code)

I'm trying to use the REST API to bring data from Mango into another system. I can authenticate and get data point information, however, I can't figure out how to get current values or historic values. Seems like I must be missing something simple in the documentation?

import requests
from requests.auth import HTTPBasicAuth

userName = 'myusername'
token = 'mypassword'
theAuth = HTTPBasicAuth(userName, token)

session = requests.Session()
response = session.get('https://monitoring.OurDomain.com/rest/v1/users/current', auth=theAuth)

if response.status_code == 200:
    print("Success Logging In")
    payLoad = {'xid': 'internal_mango_uptime_hrs'}
    response = session.get('https://monitoring.OurDomain.com/rest/v1/data-points/', data=payLoad, auth=theAuth)
    print(response.status_code)
    print(response.text)
else:
    print("Error")
    print(response.status_code)
    print(response.text)
#  log a auth fault 

It took me a little longer than it probably should have to get this up and running so I wanted to post some supplemental documentation to help other users out.

First, ensure in modules that Twilio is up to date and licensed.

You'll need a Twilio account, even a free account works for testing.

In the publisher configuration...

The Twilio account URL can be found by logging into Twilio and going to https://www.twilio.com/docs/sms/send-messages

On the right side from the languages click CURL. You'll see the HTTP POST call in the sample code, the first portion of this is the 'account URL'. In this example the account URL would be https://api.twilio.com/2010-04-01/Accounts

Your account phone number should be entered with +1xxxxxxxxxx format for a US number.

Your Account SID and Authentification Token are available in your Twilio Console.

Send To Phone Number should be entered with +1xxxxxxxxxx for a US number.

We created a watch list that select data points based on tags and its functioning exactly as we need, however, we have some usability problems with the way the CSV download file is created.

When we do CSV download separate value columns the column header is the XID which is not useful.

Also the time step of the data seems to change based on the duration of time data is selected for.

Our desired function would have to use tags to select data points. Then be able to download the data at high resolution and have the column header be DataPointName, or preferably DeviceName - DataPointName.

Any thoughts on how to get the CSV formatted like this would be great.

Thanks in advance.

@CraigWeb you're welcome. I hope this helps.

Here is a solution to the iDrive problem. This involves using their newer Linux utility. Note, I am a Linux novice so there might be more efficient ways to do some of this.

Prerequisites

1. A SSH client such as PuTTY https://www.chiark.greenend.org.uk/~sgtatham/putty/latest.html
2. A SCP client such as WinSCP https://winscp.net/eng/index.php

Configuration

1. Download the latest Linux scripts from iDrive, https://www.idrive.com/readme download the Linux scripts in the tutorial.

2. Extract the downloaded file on your Windows machine. You will see a directory called scripts which has all of the files. I renamed this directory from scripts to IDriveBackup so its more descriptive.

3. Open WinSCP.
   a. Host name is the IP address of the MangoES
   b. Change file protocol to SCP
   c. Change port number to 2222
   d. Enter the SSH credentials
   e. Click login and accept the key.

4. On the ‘right’ side of WinSCP navigate to /opt/Idrive_ARM

5. On the ‘left’ side of WinSCP navigate to where you extracted the download.
   a. Click the ‘IDriveBackup’ and then click ‘Upload’

6. If you get a permissions error you need to change the permissions for the Idrive_ARM directory. If you’re unfamiliar with Linux the easier method is through the file management tool. You can access this by going to https://[mangoIPaddress]:10000
   Select the Idrive_ARM folder then Edit -> Change Permissions.

   Change permission to 0777.

7. ‘IDriveBackup’ should now be in the MangoES.

You can close WinSCP.

8. Now connect to MangoES using PuTTY on port 2222 with the SSH credentials.
9. Navigate to the /opt/Idrive_ARM/IDriveBackup/
10. Provide an appropriate permission (executable permission) to the scripts by running:
    sudo chmod a+x *.pl

From here you can follow the tutorial on https://www.idrive.com/readme

The ‘backup’ location is the directory on your iDrive account that you want this device to be backed up to. We set this to Mangoxxxx where xxxx is the 4 digits of the particular mango device. There are steps in the tutorial to set the backup file set, this is where you set which directories to backup. This should be set to /opt/mango/

There is a utility to set the backup frequency as well.

@craigweb and @phildunlap was this ever resolved. this is the problem I'm currently experiencing

I am connected to a data source that will randomly fault and when faulted it stops responding to Mango queries via Modbus TCP and during this time Mango will continue to log new data equal to the last response from the Modbus slave. Mango logs the modbus exception in the events.

How can i stop data logging until the modbus slave becomes active again?

Below you can see where the data flat lines.

@jared-wiltshire yes that example is exactly what i'd want.

do you have guidance on a tutorial for how i write my own custom component to implement this?

I followed the advice in this post to get two watch charts on the same page and the x-axis on both gets sets by the date bar on the header, and that works great.

The challenge now is if someone clicks and zooms/pans on chart 1 the x-axis on chart 2 stays set per the date bar. Is there a method to always have the x-axis be matched on both charts even when zooming with the click and drag feature?