Replies: 4 comments 3 replies
-
I was chatting with Seth about this a bit. I'm intrigued by a new attribute I think it'd be especially interesting because I can |
Beta Was this translation helpful? Give feedback.
-
For current stats see what we log using the misc/stats.zeek script. one can New things that would be nice:
|
Beta Was this translation helpful? Give feedback.
-
They are probably included in the above notes one way or the other already, but just to call them out. |
Beta Was this translation helpful? Give feedback.
-
My comment on the event loop load was to understand how many events we process per queue drain. We have the That was almost an aside, though. In general I'd really like to see more metrics that track cluster load. I may not have sufficient understanding here (particularly with CAF's credit system) but it'd be cool to see how many messages CAF receives in each of its polls. Is that the same as Another thing that'd be really handy, but I don't know feasibility: visibility into the most frequent hits on a pool of strings/names per time unit. Picture our event names — what's are the most busy events on a given node? I guess we could approximate via one metric per event type, but that feels unwieldy. |
Beta Was this translation helpful? Give feedback.
-
With the new Telemetry feature added recently, we can now define additional metrics and telemetry data that gets reported externally. This discussion is to build a list of useful metrics to report. I've made it a checklist in case people want to start working on them, they can check them off as they're added.
Beta Was this translation helpful? Give feedback.
All reactions