You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I wonder if there is a counterpart of afl++'s rpc_statsd in LibAFL? I noted there is a prometheus monitor, but as far as I understand, this is not what I want.
The afl++'s rpc statsd has the following abilities:
A TUI monitor and a statsd monitor at the same time.
The statsd monitor sends the metrics (to a statsd aggregator), instead of host the metrics
Currently, the prometheus monitor is not good for incremental fuzzing, i.e., there may be several new targets to be fuzzed every day, so I want to add those stats as well. However, in current solution, each fuzzing instance needs a separate port, and we need to aggregate stats from all ports to show a total information.
The text was updated successfully, but these errors were encountered:
Interesting, I don't personally use either statsd nor prometheus.
I though they are roughly equivalent but looking at the code you're right.
TL;DR: I don't think we have that right now.
Should be rather easy to implement though, probably top of the prometheus monitor, or one of the OnDisk monitors.
(Personally I don't have time atm / no test setup)
And yet there is another question: can we have two or more monitors at the same time? The tui monitor and the to-be-implemented statsd monitor should both be there in my experience.
We could look at the statsd monitor for the overall-information of all fuzzing instances we are running, and look at the individual tui monitor for detail information.
You could build a simple monitor that takes two child monitors and forwards every incoming message to both.
That would be very useful to have in any case
I wonder if there is a counterpart of afl++'s rpc_statsd in LibAFL? I noted there is a prometheus monitor, but as far as I understand, this is not what I want.
The afl++'s rpc statsd has the following abilities:
Currently, the prometheus monitor is not good for incremental fuzzing, i.e., there may be several new targets to be fuzzed every day, so I want to add those stats as well. However, in current solution, each fuzzing instance needs a separate port, and we need to aggregate stats from all ports to show a total information.
The text was updated successfully, but these errors were encountered: