## Issue Addressed
Closes#800Closes#1713
## Proposed Changes
Implement the temporary state storage algorithm described in #800. Specifically:
* Add `DBColumn::BeaconStateTemporary`, for storing 0-length temporary marker values.
* Store intermediate states immediately as they are created, marked temporary. Delete the temporary flag if the block is processed successfully.
* Add a garbage collection process to delete leftover temporary states on start-up.
* Bump the database schema version to 2 so that a DB with temporary states can't accidentally be used with older versions of the software. The auto-migration is a no-op, but puts in place some infra that we can use for future migrations (e.g. #1784)
## Additional Info
There are two known race conditions, one potentially causing permanent faults (hopefully rare), and the other insignificant.
### Race 1: Permanent state marked temporary
EDIT: this has been fixed by the addition of a lock around the relevant critical section
There are 2 threads that are trying to store 2 different blocks that share some intermediate states (e.g. they both skip some slots from the current head). Consider this sequence of events:
1. Thread 1 checks if state `s` already exists, and seeing that it doesn't, prepares an atomic commit of `(s, s_temporary_flag)`.
2. Thread 2 does the same, but also gets as far as committing the state txn, finishing the processing of its block, and _deleting_ the temporary flag.
3. Thread 1 is (finally) scheduled again, and marks `s` as temporary with its transaction.
4.
a) The process is killed, or thread 1's block fails verification and the temp flag is not deleted. This is a permanent failure! Any attempt to load state `s` will fail... hope it isn't on the main chain! Alternatively (4b) happens...
b) Thread 1 finishes, and re-deletes the temporary flag. In this case the failure is transient, state `s` will disappear temporarily, but will come back once thread 1 finishes running.
I _hope_ that steps 1-3 only happen very rarely, and 4a even more rarely. It's hard to know
This once again begs the question of why we're using LevelDB (#483), when it clearly doesn't care about atomicity! A ham-fisted fix would be to wrap the hot and cold DBs in locks, which would bring us closer to how other DBs handle read-write transactions. E.g. [LMDB only allows one R/W transaction at a time](https://docs.rs/lmdb/0.8.0/lmdb/struct.Environment.html#method.begin_rw_txn).
### Race 2: Temporary state returned from `get_state`
I don't think this race really matters, but in `load_hot_state`, if another thread stores a state between when we call `load_state_temporary_flag` and when we call `load_hot_state_summary`, then we could end up returning that state even though it's only a temporary state. I can't think of any case where this would be relevant, and I suspect if it did come up, it would be safe/recoverable (having data is safer than _not_ having data).
This could be fixed by using a LevelDB read snapshot, but that would require substantial changes to how we read all our values, so I don't think it's worth it right now.
* Layer do_atomically() abstractions properly
* Reduce allocs and DRY get_key_for_col()
* Parameterize HotColdDB with hot and cold item stores
* -impl Store for MemoryStore
* Replace Store uses with HotColdDB
* Ditch Store trait
* cargo fmt
* Style fix
* Readd missing dep that broke the build
* Remove redundant method
* Pull out a method out of a struct
* More precise db access abstractions
* Move fake trait method out of it
* cargo fmt
* Fix compilation error after refactoring
* Move another fake method out the Store trait
* Get rid of superfluous method
* Fix refactoring bug
* Rename: SimpleStoreItem -> StoreItem
* Get rid of the confusing DiskStore type alias
* Get rid of SimpleDiskStore type alias
* Correction: A method took both self and a ref to Self
* Improve error handling in block iteration
* Introduce atomic DB operations
* Fix race condition
An invariant was violated: For every block hash in head_tracker, that
block is accessible from the store.
* Add LRU caches to store
* Improvements to LRU caches
* Take state by value in `Store::put_state`
* Store blocks by value, configurable cache sizes
* Use a StateBatch to efficiently store skip states
* Fix store tests
* Add CloneConfig test, remove unused metrics
* Use Mutexes instead of RwLocks for LRU caches
* Apply clippy lints to beacon node
* Remove unnecessary logging and correct formatting
* Initial bones of load-balanced range-sync
* Port bump meshsup tests
* Further structure and network handling logic added
* Basic structure, ignoring error handling
* Correct max peers delay bug
* Clean up and re-write message processor and sync manager
* Restructure directory, correct type issues
* Fix compiler issues
* Completed first testing of new sync
* Correct merge issues
* Clean up warnings
* Push attestation processed log down to dbg
* Add state enc/dec benches
* Correct math error, downgraded logs
* Add example for flamegraph
* Use `PublicKeyBytes` for `Validator`
* Ripple PublicKeyBytes change through codebase
* Add RPC error handling and improved syncing code
* Add benches, optimizations to store BeaconState
* Store BeaconState in StorageContainer too
* Optimize StorageContainer with std::mem magic
* Add libp2p stream error handling and dropping of invalid peers
* Lower logs
* Update lcli to parse spec at boot, remove pycli
* Fix issues when starting with mainnet spec
* Set default spec to mainnet
* Fix lcli --spec param
* Add discovery tweak
* Ensure ETH1_FOLLOW_DISTANCE is in YamlConfig
* Set testnet ETH1_FOLLOW_DISTANCE to 16
* Fix rest_api tests
* Set testnet min validator count
* Update with new testnet dir
* Remove some dbg, println
* Add timeout when notifier waits for libp2p lock
* Add validator count CLI flag to lcli contract deploy
* Extend genesis delay time
* Correct libp2p service locking
* Update testnet dir
* Add basic block/state caching on beacon chain
* Add decimals display to notifier sync speed
* Try merge in change to reduce fork choice calls
* Remove fork choice from process block
* Minor log fix
* Check successes > 0
* Adds checkpoint cache
* Stop storing the tree hash cache in the db
* Handles peer disconnects for sync
* Fix failing beacon chain tests
* Change eth2_testnet_config tests to Mainnet
* Add logs downgrade discovery log
* Remove dedunant beacon state write
* Fix re-org warnings
* Use caching get methods in fork choice
* Fix mistake in prev commit
* Use caching state getting in state_by_slot
* Add state.cacheless_clone
* Less fork choice (#679)
* Try merge in change to reduce fork choice calls
* Remove fork choice from process block
* Minor log fix
* Check successes > 0
* Fix failing beacon chain tests
* Fix re-org warnings
* Fix mistake in prev commit
* Attempt to improve attestation processing times
* Introduce HeadInfo struct
* Used cache tree hash for block processing
* Use cached tree hash for block production too
* Range sync refactor
- Introduces `ChainCollection`
- Correct Disconnect node handling
- Removes duplicate code
* Add more logging for DB
* Various bug fixes
* Remove unnecessary logs
* Maintain syncing state in the transition from finalied to head
* Improved disconnect handling
* Add `Speedo` struct
* Fix bugs in speedo
* Fix bug in speedo
* Fix rounding bug in speedo
* Move code around, reduce speedo observation count
* Adds forwards block interator
* Fix inf NaN
* Add first draft of validator onboarding
* Update docs
* Add documentation link to main README
* Continue docs development
* Update book readme
* Update docs
* Allow vc to run without testnet subcommand
* Small change to onboarding docs
* Tidy CLI help messages
* Update docs
* Add check to val client see if beacon node is synced
* Attempt to fix NaN bug
* Fix compile bug
* Add notifier service to validator client
* Re-order onboarding steps
* Update deposit contract address
* Update testnet dir
* Add note about public eth1 node
* Fix installation link
* Set default eth1 endpoint to sigp
* Fix broken test
* Try fix eth1 cache locking
* Be more specific about eth1 endpoint
* Increase gas limit for deposit
* Fix default deposit amount
* Fix re-org log
* Implement freezer database for state vectors
* Improve BeaconState safe accessors
And fix a bug in the compact committees accessor.
* Banish dodgy type bounds back to gRPC
* Clean up
* Switch to exclusive end points in chunked vec
* Cleaning up and start of tests
* Randao fix, more tests
* Fix unsightly hack
* Resolve test FIXMEs
* Config file support
* More clean-ups, migrator beginnings
* Finish migrator, integrate into BeaconChain
* Fixups
* Fix store tests
* Fix BeaconChain tests
* Fix LMD GHOST tests
* Address review comments, delete 'static bounds
* Cargo format
* Address review comments
* Fix LMD ghost tests
* Update to spec v0.9.0
* Update to v0.9.1
* Bump spec tags for v0.9.1
* Formatting, fix CI failures
* Resolve accidental KeyPair merge conflict
* Document new BeaconState functions
* Fix incorrect cache drops in `advance_caches`
* Update fork choice for v0.9.1
* Clean up some FIXMEs
* Fix a few docs/logs
* Update for new builder paradigm, spec changes
* Freezer DB integration into BeaconNode
* Cleaning up
* This works, clean it up
* Cleanups
* Fix and improve store tests
* Refine store test
* Delete unused beacon_chain_builder.rs
* Fix CLI
* Store state at split slot in hot database
* Make fork choice lookup fast again
* Store freezer DB split slot in the database
* Handle potential div by 0 in chunked_vector
* Exclude committee caches from freezer DB
* Remove FIXME about long-running test
* Renamed fork_choice::process_attestation_from_block
* Processing attestation in fork choice
* Retrieving state from store and checking signature
* Looser check on beacon state validity.
* Cleaned up get_attestation_state
* Expanded fork choice api to provide latest validator message.
* Checking if the an attestation contains a latest message
* Correct process_attestation error handling.
* Copy paste error in comment fixed.
* Tidy ancestor iterators
* Getting attestation slot via helper method
* Refactored attestation creation in test utils
* Revert "Refactored attestation creation in test utils"
This reverts commit 4d277fe4239a7194758b18fb5c00dfe0b8231306.
* Integration tests for free attestation processing
* Implicit conflicts resolved.
* formatting
* Do first pass on Grants code
* Add another attestation processing test
* Tidy attestation processing
* Remove old code fragment
* Add non-compiling half finished changes
* Simplify, fix bugs, add tests for chain iters
* Remove attestation processing from op pool
* Fix bug with fork choice, tidy
* Fix overly restrictive check in fork choice.
* Ensure committee cache is build during attn proc
* Ignore unknown blocks at fork choice
* Various minor fixes
* Make fork choice write lock in to read lock
* Remove unused method
* Tidy comments
* Fix attestation prod. target roots change
* Fix compile error in store iters
* Reject any attestation prior to finalization
* Begin metrics refactor
* Move beacon_chain to new metrics structure.
* Make metrics not panic if already defined
* Use global prometheus gather at rest api
* Unify common metric fns into a crate
* Add heavy metering to block processing
* Remove hypen from prometheus metric name
* Add more beacon chain metrics
* Add beacon chain persistence metric
* Prune op pool on finalization
* Add extra prom beacon chain metrics
* Prefix BeaconChain metrics with "beacon_"
* Add more store metrics
* Add basic metrics to libp2p
* Add metrics to HTTP server
* Remove old `http_server` crate
* Update metrics names to be more like standard
* Fix broken beacon chain metrics, add slot clock metrics
* Add lighthouse_metrics gather fn
* Remove http args
* Fix wrong state given to op pool prune
* Make prom metric names more consistent
* Add more metrics, tidy existing metrics
* Fix store block read metrics
* Tidy attestation metrics
* Fix minor PR comments
* Allow travis failures on beta (see desc)
There's a non-backward compatible change in `cargo fmt`. Stable and beta
do not agree.
* Tidy `lighthouse_metrics` docs
* Fix typo
