Hi
Sudarshan Kumawat
I hope you are doing fine.
I would like to know if you are referring to this
eps functionality?
By default, the manager does not apply any EPS limit, this number depends on the hardware on which the system is installed.
From 4.4 onwards, this new functionality is included, which allows limiting the number of events processed by the manager.
What is it for?
It allows to have more control over the number of events to be processed in an environment, useful for cloud developments, which usually charge for data processing.
How does it work?
The manager receives the events to be processed and they are stored in a queue, then they are processed by the decoders and rules. The limit is applied at the midpoint between the reception of events and processing, avoiding the loss of events in the case of eventual peaks. If the reception of events is continuous and greater than the configured limit, at some point the reception queue will fill up and events will start to be lost.
How to configure it?
include in
ossec.conf file the limits configuration, for exaple following configuration:
<global>
<limits>
<eps>
<maximum>10</maximum>
<timeframe>10</timeframe>
</eps>
</limits>
</global>
The above configuration means that the manager will process 10 events per second, within 10 seconds,100 events in 10 seconds. If the limit (100 events) is reached before 10 seconds, it will stop processing events until complete the 10 seconds, after that the events to be processed will be 100 in the next cycle.
In addition, if the limit is not reached, they are not accumulated for the next cycle, the limit is reset at the start of a new cycle and they return to 100.
Example: each number represents a number of events processed.
|First cycle | Second cycle | Third cycle|20 20 10 10 20 20 -- -- -- -- | 5 5 5 5 10 10 10 10 10 10 | ...............Here you can find some examples
Let me knof if that helps.
Regards.