Temp Sensor & SNMP Check for Health Status

Just got a SensorGateway with 2 additional temp sensors. All it working ... I can get the temperature from the sensors via SNMP from Nagios.

But ... is it possible to get a status on the device what the actual status is of the alarms via SNMP? There is the "LastError" for each sensor ... but that is ... well ... an error ...

I would like something like "Current Sensor Health Status" ... using the critical/warnings that is set on the device for that sensor ... is that even possible?

I have not looked at the other HTTP API ... maybe it possible there ... I just want to make 100% sure it's not something I have missed in the SNMP ...

Comments

  • @syska

    The status should also have an OID , you should be able to check it using an MIB browser and do an SNMP walk.
  • We get this:
    SNMPv2-SMI::enterprises.17095.4.3.0 = STRING: "Aircon 1,19.37,WARNING,09 June 2018,00:58:20"
    SNMPv2-SMI::enterprises.17095.4.4.0 = STRING: "Aircon 2,19.56,WARNING,01 January 2012,07:10:51"

    But it seems a little wrong ... Guess I need to get access to the web interface of the sensor to be able to determine if this is a setup error ...

    But I'm consused about the "wrong date" ... yes, the Date on the box wrong corrected later(see "Aircon 2") ... but then the date for "Aircon 2" should also have changed ...
  • Still no luck ...

    image
    This however makes me wonder ... why don't I get a recovery alert ...

    image
    This is the sonsor settings ...


  • edited June 13
    image

    WIth an "warning" in the "alert logs" this is what I get from the JSON endpoint ...

    Do I have a misconfiguration some where?

    image
    Here is the Device Info:
  • @syska

    you have to do an SNMP walk and then check for the OID that gives the "warning" string for each of your sensors. You can use any MIB browser to perform the SNMP WALK command.

    The gateway can be configured to send alerts as a stand alone. thresholds and alert types can be set under Sensor Status. But since you are using Nagios you wont get the "recover" alerts unless you configured them on your software.

    You might need to see if the date and time of your Gateway matches with Nagios.



  • edited June 21
    It sure looks to me like the a "current status" OID is missing from the MIB.

    For example:
    $ snmpwalk -v2c -c redacted -m "./sensorgateway.mib" redacted . | grep sensor3
    ServersCheck::sensor3name.0 = STRING: "Ext. Temp1"
    ServersCheck::sensor3Value.0 = STRING: "79.47"
    ServersCheck::sensor3LastErrMsg.0 = STRING: "DOWN"
    ServersCheck::sensor3LastErrTime.0 = STRING: "20 June 2018,09:40:56"
    ServersCheck::sensor3TrapErrMsg.0 = STRING: "Ext. Temp1,79.36,DOWN,20 June 2018,09:40:56"
    ServersCheck::sensor3Name.0 = STRING: "-"
    ServersCheck::sensor3value.0 = STRING: "-"
    ServersCheck::sensor3ErrState.0 = STRING: "-"
    ServersCheck::sensor3lastErrTime.0 = STRING: "-"
    ServersCheck::sensor3lastErrMsg.0 = STRING: "-"


    Also, what's up with the second set of values that are all "-"? Is it an error in the MIB?

Leave a Comment

bolditalicunderlinestrikecodeimageurlquotespoiler