Indeed a very plausable explanation. The question is if this is true.
Would it be possible to verify the IPs af these monitors?
This does not matter one bit. All that matters is that many (at least 1) marks your server as being online and keep time within limits to be right.
It does not matter where the monitor is. Really it doesnāt matter.
The pool want to know 3 things:
1: Can you be reached? If yes OK
2: Is your time response specs? If yes OK
3: Is your time accurate? If yes OK -=> 20!
You get 20 and you are part of the pool. Why should the monitor in your country?
Makes no sense, as when ISPās in your country go bad and only monitors in your country go bad, than you are out of the pool.
Do not forget that many ISPās have many peers (different routes) to other destinations.
Meaning that your ISP and another in your country may not have the fastest route, or even reliable route.
Do not trust any monitor apart from the one that scores you well, no matter where it is.
As THIS monitor is able to get your time, get you being online etc.
Or you rather be kicked out of the pool? You have to look at the bigger picture.
It does not matter where the monitors areā¦it matters they determine your server is OK and proper ticking. The rest is rubbish, sorry.
People have hobbies, you know. Things theyāre interested in. Things they wanna keep learning about. They might notice things that are different from how they would expect, and want to figure out why that is and talk about it with likeminded people, even if itās a complete non-issue.
You donāt need to correct people for troubleshooting a non-issue out of curiosity.
Where do you see that try to correct people?
I try to explain how it works and why low scoring by non-active monitors are to be ignored.
Read what I write, Iām explaining. Have you read what I wrote?
Not correcting anything. If my explenation is wrong, please say so.
Your explanation is factually correct. I simply believe you were misinterpreting what ebahapo is actually saying.
In my interpretation, heās just observing that, while there are monitors in his country, none were monitoring his server. While this is not a problem, it goes counter to how the system is believed to function, which piqued his curiosity, hence investigating. Not out of operational necessity, but out of curiosity.
Your response seemed like you were trying to shut down his investigation.
That would make sense and confirm what others reported here, that a monitor likely visits all servers in the world.
And the algorithm is mostly correct, selecting monitors in the same or neighboring countries, as theyāll tend to reach the top score for a stable server more consistently due to shorter RTT.
Yeah, I added the āpausedā mode so the system can "pauseā monitors that are extremely unlikely to ever be selected as testing or active, but right now eventually all monitors will monitor all servers.
Glad to see that your observation is that the ābestā (best-ish?) monitors eventually are selected. Iām sure itās not working perfectly, but the goal is just to have it be good enough that poor monitor selection doesnāt cause poor results.
I agree with some of the other discussion (maybe another thread?) that with the system as it is now, maybe we could have the monitors āfailā a test if any packets are lost (with a count of how many were lost in the data).