[Merged by Bors] - Detect double PoST inclusion malfeasance #6117
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #6117 +/- ##
=======================================
Coverage 81.9% 81.9%
=======================================
Files 308 308
Lines 33795 33843 +48
=======================================
+ Hits 27705 27750 +45
- Misses 4318 4319 +1
- Partials 1772 1774 +2 ☔ View full report in Codecov by Sentry. |
| db := sql.InMemory() | ||
|
|
||
| // one atx | ||
| atx0, blob := newAtx(t, sig, withPublishEpoch(1)) |
There was a problem hiding this comment.
newAtx returns a wire.ActivationTxV1 as blob. I think newAtx doesn't need to return any wire type for the tests here, if a blob is needed for a test it should generally be possible to just use types.RandomBytes(20) since the blob should not be needed to be decoded in any test in this package.
Wdyt?
There was a problem hiding this comment.
I wouldn't say I like returning a fake value when a real one is so easy to use. It's a footgun waiting to shoot somebody in the future. However, the blob could be obtained by calling atx0.Blob(), so we can only return the ATX.
I will refactor it to return only ATX in a separate PR.
There was a problem hiding this comment.
The problem I see with returning a real value is that for this we probably will need a dependency on activation and/or activation/wire but sql packages shouldn't depend on those (cyclic dependencies). Using xxx_test packages for tests is a way around this, but since sql packages can't even decode the bytes by themselves I don't see an issue with faking the bytes that are stored / returned in tests.
|
bors merge |
## Motivation With ATX merge, a dishonest smesher might try to include his PoST twice for doubled rewards: - once in a merged ATX (published by another identity) - once in a self-published ATX
|
Build failed: |
|
bors merge |
## Motivation With ATX merge, a dishonest smesher might try to include his PoST twice for doubled rewards: - once in a merged ATX (published by another identity) - once in a self-published ATX
|
Build failed (retrying...): |
## Motivation With ATX merge, a dishonest smesher might try to include his PoST twice for doubled rewards: - once in a merged ATX (published by another identity) - once in a self-published ATX
|
Build failed: |
|
bors merge |
## Motivation With ATX merge, a dishonest smesher might try to include his PoST twice for doubled rewards: - once in a merged ATX (published by another identity) - once in a self-published ATX
|
Pull request successfully merged into develop. Build succeeded: |
spacemesh-bors
bot
changed the title
Motivation
With ATX merge, a dishonest smesher might try to include his PoST twice for doubled rewards:
Description
Detect if any of the IDs participating in the processed ATX has already contributed its PoST to another ATX in the epoch
💡 This PR doesn't implement the malfeasance proof, just malfeasance detection. The proof is stubbed and will come in a later PR.
Test Plan
added tests
TODO