One of my nodes went offline, and I couldn’t reach it phisically, it has a bad system drive that I needed to change. It took me about 3 weeks to getting it back online (03.29). The data had no intact. It is online in the last 2 weeks (04.17. right now). I got 1 unsuspended and 2 disqualified emails in the last 2 days Why we can’t see the online windows, in the different sattellites? Why it is not public data? Why there is no communication before an automatic disqualification? (Why we don’t have domain names in the online/offline emails?) I have 10TB of data, it was online more than 2 years (maybe closer to 3). It seems really unfair to me, and seems to be a joke.
I will check the satellite/data ratio and maybe keep it online, but I’m close to “unplug” my other server too.
You should have received an email when the node was detected offline. Then when it hit 60%-uptime there should be a second email saying it was suspended. And finally if you don’t bring it back in time… another email saying it’s disqualified.
Those emails tell you the Node ID with the problem (and not the IP, or hostname) because the ID is unique. An IP or hostname could be recycled by the time you can address the issue.
Some would say having a node offline that long… while receiving emails telling you about it… and still expecting to be paid… would be the joke . It’s unfortunate: accidents happen: luckily there are almost 28000 other nodes online to pick up the slack. And it’s free to roll a new Identity and try again!
I found the offline process a bit wonky, but in the opposite direction. I had a node which was down for well over 3 weeks, and only two satellites (IIRC, US and AP) noticed. Honestly, if there was also a bug where a satellite would not notice a node back up, I would understand grievances. @Roxor’s message is unnecessarily critical here.
I would recommend to check the reason of disqualification. The disqualification for being offline is happening only if your node managed to be offline more than 30 days.
So, It’s likely the node corrupted pieces or lost them.
And finally if you don’t bring it back in time… another email saying it’s disqualified.
Yapp, exactly, it would be “nice” to get a 2nd email after 1 days, and after that at least weekly a “you are still offline” email bcs sometimes the system sends online-offline emails 10 minutes apart from each other. And they come from different satellites and different times, and its a total mayhem.
An IP or hostname could be recycled by the time you can address the issue.
It would be a “nice” UX if the email would contain it besides the id. It would cost nothing but would give the other and a lot faster way to understand which site is the problematic.
while receiving emails telling you about it
You mean receiving ONE email about it?
and still expecting to be paid
I’m not expecting to be paid. I’m totally ok to loose 1 month of income if my node went down for 2 weeks. I think THAT is fair. I think loosing all of your progress, and starting it from 0 is unfair (and probably unhealthy for the network). It’s not like a skateboard jump that you can just try again and get better with it after every fail.
Long story short: I think the node operator experience is really poor. Maybe its my fault to get disqualified, I can accept that, but I can’t say that “its entirely my fault, and the whole process cannot be upgraded significally”.
One for offline, then one for suspended. Two email while it’s counting-down to disqualification. At least that’s what I receive.
If you expect anything other than starting-from-zero: then you’re “expecting to be paid”. The node disappeared for weeks (3+?) doing nothing useful for the network. Storj needs to maintain a healthy number of pieces: they can’t hope-for-the-best when a node disappears so long. And if all the data you held is covered by other nodes by the time yours eventually/maybe comes back online… it should be paid for the zero-useful-data it now holds. That means $0.
But new Identities are free, so it’s easy to start again!
The network is built for nodes to be disposable. So quick repairs are what’s good for the paying customers. Waiting forever for offline nodes who may-or-may-not-ever-come-back would be unhealthy for the network.
Why we can’t see the online windows, in the different sattellites? Why it is not public data? Why there is no communication before an automatic disqualification? (Why we don’t have domain names in the online/offline emails?) I have 10TB of data, it was online more than 2 years (maybe closer to 3). It seems really unfair to me, and seems to be a joke.
. It’s unfortunate: accidents happen: luckily there are almost 28000 other nodes online to pick up the slack. And it’s free to roll a new Identity and try again!