Join GitHub today
Script to collect port polling data and compare full walk vs selectiv… #7626
…e port polling.
DO NOT DELETE THIS TEXT
If you would like to test this pull request then please run:
referenced this pull request
Nov 4, 2017
Works for me. Some comments:
Add support for specifying hostnames / wildcards?
Show what the values are in (seconds) for the time column.
Change color in difference column to show if the change would be a bonus (or indicate that a - in that column is a good thing).
Flag to auto enable selected port polling for devices it's worth it for?
Added to performance docs.
I know it's a lot to ask ^^ but this quickly shows where you can get some good savings.
I'm glad you guys are excited :) I just threw this together as a quick way to get data so we could make a better decision on automatically enabling per-port polling.
Off the top of my head, so might be slightly wrong :D
I managed! I think I'm starting to understand git now.
I added all the features in our comments, but I'm not very happy with all the hoops I had to jump through to get the printf work correctly with colors. There are also docs and CLI help to be added.
But I'll have to leave that for now, will work on it more on Sunday or Monday. I have a party to plan and host. :)
Hmm, I see; thoughts on including hostname? Since we are now intending people use this to set up per-port polling should we include hostname in the output?
I originally excluded it because I just wanted people to send us the output so we could make an informed decision about enabling per-port polling automatically.
I think we can definitely allow people to make more use of this with the extra features. Trying to use it to gauge if we should turn it on globally may not give us enough data to make that decision.
Would be better off for us to start tracking some of the data in polling so we can record max time for a full port walk vs selected port polling and switch from one to the other as we need to maintain performance.