Hi, I’ve recently created a 8.4TB Storj node and have just worked out that at this rate I’m looking at 6-7 years to fill the node up. Does this seem correct? The Storj Earnings Estimator suggests 0.25TB in month 1, but in the four weeks I’ve been running I’m at 29GB storage + 45GB gone to trash.
I have 1000/1000 fibre connection which speedtests at 900/700 so I wouldn’t expect that’s the issue, and uptime has been good. The Storj node is running on a TrueNAS app.
Any advice what to check to improve the fill rate?
I’d start off my saying, that StorJ is a turtle race, it’s never going to be fast. That being said, 29GB in a month does seem small. It could be that your node has not fully vetted yet. In that case, more data will come. It could also be, that there’s an issue with the node.
What is your ping to your ISP?
Are you running an SMR harddrive?
Have you checked if you have any neighbors? Neighbors
I totally understand it will be slow starting off, it just seems very slow and using approx 0.4Mbps of my internet connection. Would location be a factor being in New Zealand and further away from the Storj servers than other users?
I think the upload may be being throttled right now as it’s evening / peak time.
The hard drive is a 10TB Seagate Irowolf (ST10000VN000) which uses CMR technology.
Neighbours looks good, only showing 1 node in the subnet.
I am only forwarding port 28967. The TrueNAS Storj deployment guide (Deploying a Storj Node | TrueNAS Documentation Hub) did specify some other ports but this guide seemed outdated and I think followed the Storj docs more than the TrueNAS guide.
I have a static IP too, so no DDNS service required.
Low ping, non-SMR hard disk and only one node in your subnet.
June has been slowish for me as well, with a lot of deletes going on, but I’d point fingers towards your vetting status. You’re out of luck, because in a few releases, vetting status will be visible on the dashboard, until then, you’ll have to run the vetting-checker script by hand.
Filling around 2TB/year sounds reasonable, lately. Like Ottetal said: it’s kinda like watching paint dry: you’re better off ignoring it. The system will send you email if it thinks your node has gone offline… so once you’ve set it up just go out and live your life
Ping to customers is what matters. Most customers are in the us, some in europe, almost none in your region. So new zealand is probably not the best spot to run storj nodes.
I would add that there are actually customers in New Zealand and Australia. However, the largest ones use AU Select due to SOC2-like requirements.
But you still have to get downloads from people all over the world, your nodes can win a lot of races while the closest nodes to customers in other regions are full.
Your 10 TB drive will take > 5 years at this rate to fill, and one year or so left of life thereafter… Maybe re-consider your ROI at that location, or wait a month after being vetted. I say this as it sounds as though you’re committing this space to this specific purpose.
This is available in the Storj Console (satellite web UI) or in the linksharing service.
In the Storj Console it’s a small globe icon in a preview screen, on linkshare it’s available on the left side.
There is also a parameter "?map" at the end of the linksharing URL to see only the map.
As @alpharabbit pointed out, ping to customers is also imported. I think you’re getting hit with a double whammy: You’re not vetted on some of the satellites, which severely hampers the amount of traffic sent to you. You’re then not getting a lot of vetting audits, because you’re far from most customers, thus lengthening the vetting process.
If I was you, I’d be running the node on hardware that’s online either way. With a bit of luck, KiwiNation will start to use the service more and more in the future, in which case you have a head start with vetted nodes
