we have a new champion which survived churnmagedon dial-up.mp4
ant file download --retries 20 b1f02ea6fe81c4d9bcdf62f4b5bee9b0852ec5e81adf0569b7bc738125975962 dial-up.mp4
I have now added it into the testing script.
we have a new champion which survived churnmagedon dial-up.mp4
ant file download --retries 20 b1f02ea6fe81c4d9bcdf62f4b5bee9b0852ec5e81adf0569b7bc738125975962 dial-up.mp4
I have now added it into the testing script.
dial-up has just been deposed rough night by patosh has survived since the genesis event and was uploaded at same time as beg blag was original uploaded.
ant file download --retries 20 720fb8ebfaae59d22516c6091ce99c482ebe255f5bb019430109e528f620bfc9 Patosh-RoughNight.mp3 #md5sum 452d1231d72489503ce73bc504b3da6e 6Mb
It looks like my nodes are now storing more like 200-500MB, rather than 5-10MB, so definitely a big change (about 50x more data).
Obviously, 500MB is still some way off 35GB, so there are still must be plenty of nodes vs amount of data to host, but it’s moving in a positive direction.
I can also see the rewards creeping up. I’ve seen 12 payments over the last 24h, starting at 0.0000003 and ending at 0.000003 (i.e. 10x more).
think we have golden nodes again out my 100 nodes one node has had 9 payments lucky node 37 the rest zero
I would not trust that ever since the 1 to 1 updating of the field was gone from the code. It is more random now if a node is even told that it was paid or not. Also does Merkle record it as one or the total number of records in the tree?
So it might be one payment and that was for 9 records in the tree.
Just looking at three machines each running 100 nodes, 2 of them have 1 node being paid each with one at 13 payments and one at 8 payments. Seems to me that rather than “golden” nodes it is more like the number of payments is the number of records in the tree being paid for. Just random chance your node was the one chosen. The 3rd machine has 3 nodes being paid but no amounts reported by /metrics.
EDIT: and looking at the blockchain I cannot see that many payments for any of the accounts (1 per machine). Also the amounts do not match up at all. Metrics might be telling porkies as well on this metric
they came in from 11:30 to 14:30 can see the line going up in increments.
yea not sure whats going on but at least rewards are coming in ![]()
I think merkle payments are made in one go that payment was for 0.00000961 which is about 4 x all my other payments.
I just got my first 5 dp reward: 0.00001266174643681
Prices are definitely responding nicely to the increased storage used per node.
Nice to see those noughts disappearing for the better for a change ![]()
Something I like seeing is the ratio of responsible records compared to total records stored is increasing. It was like 0 or 1 responsible for 20 to 50 records stored for a node. Now its like 60 to 140 responsible to 200 to 300
Bandwidth still seems quite high which suggests to me that there are still a lot of nodes going down and then node runner is starting new ones. Hopefully if someone has their nodes die that they start again with a reduced number of nodes and that way they have a much better chance of their nodes staying up. Also in time will reduce bandwidth on every one
downloads are significantly faster check the times from the last run we have gone from 30s to download to 1 or two seconds ![]()
Thats because I am back up and running
![]()
Ironically there is probably some truth to that. A different class of node runners are left operating now. ![]()
And they are not getting overwhelmed by leeches.
yea at least emissions has given the team plenty to think about with regards to what people will get up to.
Do you think this is because of the smaller network or because the nodes are “better”?
better nodes is my guess