@Walrus 🦭/acc
In Walrus, during each epoch, storage nodes get hit with random challenges for the blobs they're supposed to hold.
It triggers when the system picks a specific blob at a pseudorandom moment (tied to an on-chain event), and the node has to quickly put together and submit a compact Merkle proof showing it still has its sliver intact.
If the response doesn't come through on time, the node fails the availability check, and that gets recorded on Sui.
WAL
--
--