{% extends "base.html.twig" %} {% block title %}How to use Agent Profiles{% endblock %} {% block body %}

How to manage Agent Profiles


The purpose of agent profiles is to be able to configure agent settings on a device in an automatic way.

Every agent is assigned a profile and the profile configuration parameters are checked on each data push from the device. If the parameters on the agent differs from the values defined in the profile, the agent is immediately provided with actual values, resulting in immediate update to the correct values.

The meaning of profile settable values is this:

  • Name ... just a name of the profile. For system profiles, you cannot change the name.
  • Ping ... interval (in seconds) between ping attempts. If a ping fails, the agent waits this number of seconds till the next ping attempt.
  • Regular ... number of seconds between data pushes. This is probably the most interesting parameter. Increasing this, the data updates from the device agent will happen in longer intervals, thus making the data consumption smaller but also less frequent, meaning the data you are looking at will be updates less frequently and also device reacting slowly on commands. NOTE that decreasing this to too low number has impact on higher consumtion of resources not only on device, but on the server as well (especially if this concerns a lot of devices). Also, decreasing this to a low number (below 60) may have impact on Webtunnel setup - if a device is not fast enough to finish tunnel setup between two data pushes, the tunnel setup may fail - in that case, first increase this number and then retry Webtunnel setup.
  • Debug ... this is an interval (in seconds) used to wait between the data pushes when the device agent is set to debug mode. Currently (v. 0.8), this value is not propagated to the device agent.
  • Fallback ... this interval (in seconds) is a fallback interval between data pushes that is used when all the ping attempts (3) fails consecutively. Once a data push is successful, agent switches back to the Regular interval.
  • Small Cycles ... number of data pushes (called small data pushes or small cycles) which are followed by a big cycle (big data push). Some data metrics, especially these that could consume more resources on the device to gather or consume higher amount of data during transfer, are gathered on the device and transferred to the server only during the big cycle. Lowering this increases the frequency of these big data pushes. Increasing the number lowers data and resources consumption, mainly on the devices.
  • Fail Tolerance ... the number of failed connections (connection fails after 3 unsuccessful pings) for which the data gathered are still considered fresh. Once there are more failed attempts than this number, the data are gathered again prior to next push attempt.
  • Devices count ... not an agent parameter but a counter showing you how many devices are using this profile.

Every new device agent is automatically assigned the DEFAULT agent profile.

You can assign any of the available profiles in two ways:

  • In Device Details - to the one device you are viewing
  • On the Devices page, using the Assign Agent Profile mass action button below the table. This way you assign the profile to all of the selected devices. If you want to see which Agent profiles are the devices using directly in the table, go to System -> Settings and add the field Agent Profile in the Devices Table Columns. Once you add the field, you can use it for filtering/sorting purposes as usual.

You can add any number of your own profiles and edit them at any later time. You can also delete unsued profiles you created.

As for the predefined (system) profiles (DEFAULT, FASTEST, SLOWEST), you can edit the values as you wish but you cannot delete those profiles or rename them.

{% endblock body %}