How to handle with outage

Hey all :slight_smile:

So here is my problem. The electricity company is going to do maintenance 15december. This never happened but This means my house does not have power for almost the whole day. My fiber 500/500 and everything else will be down. Is it worth it to pick everything up and move it to a datacenter for a week. My downtime will be 3 hours then. Or shall I just wait the 8/9 hours till power is restored? What are your guys thoughts ? It a new node just running 1 day now. Don’t want to do anything stupid wanna be a good node :stuck_out_tongue:

Your node is only 1 day old? At this point I wouldn’t bother doing anything if your power is going to go out worse case you just start another node simple as that. Don’t go wasting money on a VPS for 3 hours of downtime.

Ow not a VPS I can move it to rackspace if needed. But if I understand correctly it doesn’t really matter because it is a new node and I do not have any rep. Yet.

Yeah exactly your node doesnt have any rep, so if your node gets DQed for any reason there is zero loss. Maybe just maybe I would buy a UPS before buying a VPS. just sayin.

Is there a quick way to see if your node is DQ?

Yeah on your dashboard it would say it.

Also Thinking about the downtime you would need to move it and the cost of using a rackspace… Not worth it for a 1 day old node.

you can have 288 hours of downtime per month currently… so it’s nothing to worry about really…

So many hours :open_mouth: I thought it was 5 haha! Oké then I just leave it where it is. Shut it down in the morning and boot it up in the evening. Thanks guys

that’s the end goal, or what storj labs would like people to keep it to…
but i’m sure in that there will also be room for extended downtime, in that so long as the avg is maintained, even amazon aws datacenters have downtime, as demonstrated a week or two ago when most of the northamerican accessed websites run on aws was inaccessible because aws added so many computers to their datacenter that their network or whatever errored out… and took the better part of a day for them to figure out what was happening.

so 5 hours pr month is unreasonable for basically anyone, storj labs would also have to accept that natural disaster and whatever can take down large portions of nodes which will then have to be allowed for extra downtime atleast in a short period.

so yeah long story short… they want 5 hours per month, but current also accept that is a near impossible goal which they will have to figure out a way that works for everybody.

so currently its 288 hours and then a node gets suspended…

however my main node lost internet for 9 hours and that resulted in 120gb of data got deleted, so it’s not like extended downtime is free… the network will ensure data is available and if it considers you a risk, data will be moved to other places, i’m sure it’s much more complex than that… but thats how i understand it… you might also be lucky that it doesn’t change anything at all…

my main node is 14.5 TB so not really a bit hit… less than 1% so doesn’t really change anything.
so atleast for now, aim for 5 hours of dt avg pr month, but less can work.

if you hit the 288 hour downtime in a month, your node will enter suspended mode, which means it will not get any more ingress, it will still be able to earn money from egress i believe.
after 1 month of stable operation the node will go out of suspension, if the problem persists the plan is that it can get Disqualified / DQ but not sure if that part is active yet… nor how long that takes.

Great thanks then I will not worry too much :slight_smile:

Well keep in mind you also can get DQed as a new node for 1 failed audit if its your first audit.

2 Likes

might be a bit like being struck by lightning while golfing… i suppose it happens… but not enough for people not to get keeled by it… lol

not that one would get killed by lighting… at most it usually just maims…

i did forget about that tho… the 1st audit dq