Poor download levels this month (Nov 2020)

I’m having very low download levels this month - almost as fast as my uploads are happening I’m getting deletions so I’m getting less than 1GB net uploads a day (despite having plenty of free space). I’ll be lucky to get 2/3 of the download traffic I had last month.
Anyone else noticing this uploading/deletion with fewer downloads pattern?

1 Like

Welcome to the forum @waistcoat!

I hope you know this isn’t mining. You can’t expect or compare any day’s/week’s/month’s traffic with any other day/week/month. As long as you are online 24/7 and not failing audits you are good.


I’m well aware this isn’t mining…

Sorry my forum account is new, I’ve been read-only on the forums and haven’t required one until now.

I’ve been an SNO for 7 months and hope that eventually I might at least subsidise some of my costs. I know I’m unlikely to make anything substantial, however little by little my payments have been increasing, but with downloads being the major component of the payments I’m a bit surprised at this months projected figures (it’s worse than I originally thought).

Jun 49GB
July 115GB
Aug 217GB
Sep 165GB
Oct 210 GB
Nov (projected) 49GB

Looking at my logs I can see some new upload/delete/never download patterns that I haven’t noticed before. I’m not seeing many more 'download cancelled; log entries than before so I don’t think I’m losing out on latency. Am I just being unlucky or is anyone else seeing such a change this month?

I’m a tiny SNO only holding just over 1.8TB at present.

1 Like

Same for all of us:

You could also have a look at this thread where SNOs share their observations on how the network behaves:

1 Like

oh i’m having fun… like 2 months ago i discovered another node on “my” subnet, found out he had been there for atleast a month, maybe with dynamic ip because he seems to be vanishing sometimes…
so initially i thought he had left… ended up finding out he hadn’t a month later, at this point i had made an extra node which i was planning on moving to a different subnet… but the technical side of it made it delayed by quite a bit… so here i am 3 months later basically haven’t grown the storagenode in atleast 3 months… at one point it was literally shrinking…

so yeah i think i can say we all hope all this re configuring or whatever it is will be done soon…
fingers crossed just maybe ill get a new isp and better options in a weeks time…

i started my node 9 months ago and i got 14.6 tb the 14 tb i already had around the 6 month mark, i was lucky to start at an opportune time, this has thus far mainly been an experiment in viability…
and i’m sure there will be more data coming in the future, storj labs are just busy with other stuff for now i guess… can’t be sending petabytes of data every month :smiley:

Honestly I can’t see how they could keep sending data our way forever.
This will have to stop eventually one day, and data should come from customers.

Let’s just hope that “day” isn’t today… Because clearly customers aren’t very active yet.


I understand. Also it’s the holidays so customers might not be using the network a lot.

Yea, it took me about 4 days to recover from that massive delete dip too.


Out of interest, how did you scan your /24 subnet for another node?
I’d quite like to do that with mine to see whether I have competition or not… :slight_smile:

used this, been quite happy with this little tool.
tho keep in mind it will only work so long as the other node exists on a default port, else it becomes a lot more difficult to track down. (atleast by conventional means, stuff like packet sniffing)
port scanning is nice and simple…

i think there is a site for it today also… i should have the link … somewhere :smiley:

haven’t tried this out yet, as i already know there is someone on my subnet and if he has more nodes now it’s just going to annoy me, so i’m just going to not look lol…


@ACarneiro you should try @SGC’s suggestion:

I tried it myself and it did find all of my nodes.
Pretty handy feature! :slight_smile:


Yea, just tried it. Thank you guys :slight_smile:
Only my nodes showed up so I’m pretty pleased about that :wink:

the mathematical chance of having a neighbor node, is very low…

but then in some cases it’s not really random, but a small ISP having a small block of IP addresses they maybe filling up 1 IP/24 block at a time… thus if two people with the same ISP lease a fixed IP address then they may be allocated into the same IP/24 block.

so depending on the ISP approach and just random chance your odds can be quite high and insanely low, since there are like 10 million or more different IP/24 subnets used on the internet …

so with like 10k storagenodes or less then the odds are like 1/1000 … if it was random
and ofc that would only be for the last one… everybody else would be less chance :smiley: even if only a little bit :smiley:

I’m still amazed about how Germany has more nodes than the USA and all the other smaller European countries have lots of nodes too. So probability of multiple nodes on a subnet is certainly bigger in those.

1 Like

I never really understood why it is better to scatter data among random /24 subnets rather than simply fully random IPs…

I don’t clearly see why /24 subnets better spread data (supposedly hence reducing chances to concentrate data in a geographical area).

because they want it to be distributed over a large geographical region… there are much fewer IP/24
the real mistake they made was inform people on how data was distributed…

remember the number of possible IP addresses x255 times the number of IP/24 subnets… which is 16.5million so that gives you 4228250625 possible node addresses… and some people maybe have access to millions of those… at their jobs… so you could take even a small scale data center and just create nodes out of it, thousands or even millions of nodes in one place.
this way they know that when they reach like 10mil storagenodes the entire ipv4 will be filled…

so it makes good sense… because they can atleast in theory filled the entire range ipv4 range, would make them like 1000 times larger than they are today… but still it’s possible…

while them becomming… 300000 times larger is slightly unrealistic… because we get close to those astronomical size numbers.

even 1000 x is a bit of a stretch…

Until this month, I was used too seeing 500GB-1TB, are others seeing this type of drop off? Nothing about the speed or availability of my node should have changed.

1 Like

In the last day or two of November my levels seem to have picked up somewhat, although it’ll still be a leaner month than had been usual. Noticed more repair traffic than usual too.

Egress seems to have picked up a bit in the last few days for me as well.

1 Like