Mine is also appearing as offline, although that I’m passing the ping:
- started
-
- via: St.Petersburg, Russia
- QUIC: dialed node in 87ms
- QUIC: pinged node in 77ms
- QUIC: total: 164ms
- TCP: dialed node in 149ms
- TCP: pinged node in 75ms
- TCP: total: 224ms
-
- via: France (proxy, no QUIC)
- TCP: dialed node in 366ms
- TCP: pinged node in 99ms
- TCP: total: 466ms
-
- done.
But my question is more oriented towards how nodes are counted on a /24 subnet.
It is to my understanding that traffic is divided between all nodes part of the subnet, and a check that I have ran under https://storjnet.info shows that I have 2 nodes, when I’m only aware of one:
ID: 1goJaKCBNoXAcojEDrXqQDm4NJYRGEaPjjDadysXZSeT3syJJH
IP address: 194.50.122.188
Is my watchtower counted as a node running in Docker?
Could this negatively affect the traffic and furthermore the statistic overall for my node, having in mind the division of traffic?
Hello.
Thank you for the question.
/24 mean then 194.50.122.1-254 all nodes in this space is like 1 node, all traffic will be shared between them. so your node will get only half of normal data size, and will fill longer.
If you are only running one node in that /24 subnet, then the other one belongs to someone else who is also running on that same subnet. Each node will only get about half the traffic as they would if they were by themselves in the subnet.
@heunland can you clarify if this “other” node is related to mine or that would be just another user/client who is using the same IP address subnet as me?
watchtower is just a way to update, it is not a node. The other node currently operating on the same /24 subnet as yours is run by some other node operator who happens to be on the same subnet. They may be unaware of the fact that they are not the only node in the subnet.
Some other users that have noticed they are sharing traffic with another node they are not running themselves have commented that they could start additional nodes on the same subnet so that in the end they get a bigger percentage of the total traffic for that subnet, but if the other node operator does the same then the advantage would disappear again.
Storj wants to split traffic among all nodes inside a /24 subnet to geographically spread data better I guess, but it feels a bit unfair for SNOs who happen to share their traffic with an other unknown one
Naïvely removing it without any alternative, yes, it would. A single SNO would then be able to get a lot of traffic by creating many small nodes, as opposed to few bigger ones.
Yeah sure, I should have made it clear that I thought all nodes behind a single IP should still be considered as one single node. I meant to only remove the grouping per /24 subnet.
Very interesting post you linked! I’ll support it!
Yes, that is a good question, this also limits the number of nodes per country, to a very small number for some - under 10k. Whether this is good or not will be seen when Storj expands sufficiently, or some yet-unknown limitation presents itself.
This is a third party website, separate from Storj as a company/business. If you keep your node online then it should be included in the figures on that website.
Where?
On that site? If so - you must use the proper NodeID, address and port. It performs checking using the Storj protocol. This is not a simple port checker or ping, so the proper NodeID is an integral part of these checks. At least author promises that.
- via: St.Petersburg, Russia
- QUIC: dialed node in 85ms
- TCP: dialed node in 149ms
-
- via: France (proxy, no QUIC)
- TCP: dialed node in 344ms
Seems all good, I guess - but I’m not able to get the right signal if I go here, it has been marked in red since I initially added the node here: “My nodes”
