Split up StatusDeque into different modules
* LastIdQueue tracks last_ids
* StatusCache keeps track of signature statuses
* StatusCache stores success as a bit in a bloom filter
* Overhead for 1m Ok transactions is 4mb in memory
* Less concurrency between the objects, last_id and status_cache are read and written to at different points in the pipeline
* Each object has its own strategy for merging into the root checkpoint
* Use vote signer service in fullnode
* Use native types for signature and pubkey, and address other review comments
* Start local vote signer if a remote service address is not provided
* Rebased to master
* Fixes after rebase
* Also implement more storage contract logic
* Add transactions for proof validation,
* Move storage state members into system storage account userdata
* Break out last_ids into its own module
* Boot SignatureNotFound from BankError
* No longer return BankError from LastIds methods
* No longer piggypack on BankError for a LastIds signature status
* Drop all dependencies on the bank
* SignatureStatus -> Status and LastIds -> StatusDeque
* Unstable tests, issue 2193
* Remove logging init from storage program: saw a crash in a test
indicating the logger being init'ed twice.
* Add entry_height mining proof to indicate which segment the result is
for
* Add an interface to get storage miner pubkeys for a given entry_height
* Add an interface to get the current storage mining entry_height
* Set the tvu socket to 0.0.0.0:0 in replicator to stop getting entries
after the desired ledger segment is downloaded.
* Use signature of PoH height to determine which block to download for
replicator.
* Move more of the replicator logic into the replicator class
* Add support for the RPC interface to query the storage last_id value
that the replicator would sign and use to pick a block.
* Fix replicator connecting to gossip and change test to exercise that
scenario.
An RPC client that fetches the signature status before the bank finishes
executing the corresponding Transaction should receive SignatureNotFound
instead of Confirmed
* Cluster Replicated Data Store
Separate the data storage and merge strategy from the network IO boundary.
Implement an eager push overlay for transporting recent messages.
Simulation shows fast convergence with 20k nodes.
Greg Fitzgerald
Sagar Dhawan
Tyera Eulberg
Michael Vines
Tyera Eulberg
anatoly yakovenko
Stephen Akridge
Pankaj Garg
Rob Walker
Jack May