Large amount of recovery

I am extremely surprised by the amount of recovery traffic. I store 100gb in the cloud and today the repair is already 7gb or 7% for 14 days!
It turns out that 7% of the nodes died in the network? According to official statistics, which we can’t check because it is carefully hidden (as well as the roadmap :smiley:).
What do you think about this?

Customers are not getting charged for repair traffic and don’t see it in the UI. Did you get that number from the advanced report? We have a bunch of bugs open for that one. I would not trust the numbers you see there.

I could query the repair traffic in the database for you.

I will really happy to get this stats for repairtest bucket from projectID=d2adb8af-be11-450f-92eb-17715cc18312 for dec, jan and feb.

In December there was about 7.7/9 GB (up and down. I don’t now which one is which)
In January 17/18 GB
February 8/10 GB so far. Should end with 16/20 GB at the end of the month

So that looks normal to me. Is that matching what you are seeing?

1 Like

Yes, I see the value less than in reality.