Description
Why
We have noticed that there exist on the number of the Cardano stake distribution which is certified: we sign and compute the artifact on the wrong stake distribution snapshot.
After investigation, it appears that the artifact should be associated to one more epoch than the one which is announced.
Analysis
On preview
, at epoch 668 we expect to sign the stake distribution which has been computed at the end of the epoch 668 (i.e. at the very beginning of the epoch 669).
We focus on the pool pool1leml52hm4fcp3hhe4zye08qz27llhj7d339p3gs0tl85cstx59q (or fe77fa2afbaa7018def9a889979c0257bffbcbcd8c4a18a20f5fcf4c). When we query the stake distribution with the Cardano cli, we receive this information:
"fe77fa2afbaa7018def9a889979c0257bffbcbcd8c4a18a20f5fcf4c": {
"stakeGo": 1426146236321,
"stakeMark": 1427994877602,
"stakeSet": 1427131764516
}
This means that the stake distribution which was signed at the end of epoch 667 (i.e. the stake distribution of epoch 667) (the Mark snapshot which is the most recent snapshot of the stake distribution on the Cardano node) has 81427994877602 Lovelace* for the pool we are focusing on. This the value that we expect to download when we download the stake distribution of epoch 667 with the Mithril client.
When we download the stake distribution of epoch 667 with the Mithril client, we get this result:
"pool1leml52hm4fcp3hhe4zye08qz27llhj7d339p3gs0tl85cstx59q": 1427131764516,
This means that we retrieve the stake distribution for the epoch 666.
What
Fix the computation of the message and the artifact of the Cardano stake distribution so that it targets the correct epoch.
How
- Investigate the problem
- Fix the computation message and artifact with the correct epoch