* [WIP] aggressive peer connections (and logging)
* get peer list from each connected peer when low on peers
* cleanup the try_read() calls and log consistent warning messages
(will clean these up later)
* cleanup error logging (log at debug level)
* simplify sync process further - one thread, try_read on peer for robustness
* add most_work_peers() for convenience, add most_work_peers count to "monitoring peers" log msg
Added support for peer banning on the p2p server. The peer status
is changed and the peer is disconnected. A banned peer won't be
able to reconnect as well.
Tracking of chain errors due to a block that's intrinsically bad
and banning of the peer that sent it. If we're syncing, resetting
the header chain to the same as the main chain to force
backtracking.
* port simple_sync across from testnet1, sync head & improved fork handling
* introduce new sync_head for tracking header chain during sync
* add debug logging in body_sync for state of various heads
* update header_head during sync if header is now the one with most work, pass in sync and head contexts to pipe
* port across sync_head changes from master, put the 1s sleep back in...
* let sync run even if the full 512 headers are all known, give sync a chance to sync against a fork that forked a long time ago
* handle height jumping during a reorg
block validaton should not check for height+1 based on head
no assumptions should be made about height indices
* quick(er) check for previous block in process_block (we cannot check height against ctx head)
* make body_sync a _lot_ faster by finding the forked block more efficiently...
* fix monitoring peers log msg
* fix chain tests
* fix grin tests - we were using the wrong genesis hash (wrong chain type)
* apparently needs setting in both places...
* body -> header -> sync (#414)
* port over body -> header -> sync changes from testnet1
* port across "simple_sync" changes from testnet1 branch
Cleanup direct refs to peer map or peer store
P2P server acts as a facade, handling the list of connected peers
and the storage of their information. Everything else goes through
the p2p server instead of having a peer map reference or going
straight to the store.
Fix p2p tests
* fix "monitoring peers" log msg (use connected_peers)
And error in the sync run loop would just exit it silently. Abort
with a panic instead.
Peer errors (like disconnect) on locator request was bubling up in
the loop. Just logging instead.
* add total_diff to ping/pong msgs
debug log for total_diff on each ping/pong
* expose peer addr to the handle_payload fn
so we know where it came from
* fix p2p tests for ping
* default to 0 if we cannot read total_difficulty
* updating a connected peer in place
* actually update peer info diff
* fixup p2p tests
Fixes#281 by doing:
- "Starting validation pipeline for " -> "Processing "
- stdout logging default = Warning
- ERR -> INFO for "Transaction rejected: Already in pool" -- only for netadapter transaction_received, which (hopefully!) only gets tx from peers, and lots of those will be duplicates, that's good and not an error.
* Downgrade ERR even down to DEBUG.
Also let Transaction rejected show tx hash,
so users can check if such a rejection is for the
payment they expected, or just any random peer
sending in some old and already known tx.
* use a short timeout when opening the connection to possible peers
lots of peers in the db are unavailable and we do not want to wait 60s for these
* also randomize peers so we don't always try and connect to peers with low ip addresses
* bump timeout up to 5s for peering
Renamed mining parameter mode to chain type, with existing types
of CI testing, user testing, testnet1 or mainnet. The public
chains (testnet1 and mainnet) come with their fully pre-defined
genesis block.
Still need to set the nonce and cycle for testnet1 genesis.
* bad case of trying to rewind to block header height 0
* rewind_to_genesis appears to work
* do not assume genesis block at height 0 is empty, pass full block in to rewind, check for last output and kernel, use index 0 if block is empty
* use HashMap internally for tracking connected peers (to avoid duplicates)
* reuse the handshake on the server so we can track our own nonce to avoid self connections correctly
* make sure we start up the clean_peers loop even in seedless mode
* logging in monitoring peers loop
* simplify monitoring for no seeds
* fixup and cleanup simulnet tests (real seeds in places)
* only start the sync if we have either seeds or peers that we know about, exit syncer safely if we have no connected peers
* default wallet receive to listen on 127.0.0.1, optional flag on wallet command to listen on 0.0.0.0
* fix simulnet for new wallet port/interface config
Moved the HTTP APIs away from the REST endpoint abstraction and
to simpler Hyper handlers. Re-established all routes as v1.
Changed wallet receiver port to 13415 to avoid a gap in port
numbers.
Finally, rustfmt seems to have ignored specific files arguments,
running on everything.
* mount v2 router for flexibility, wallet checker now refreshes multiple outputs via single api call
* fix the api router
* wallet api handlers, miner uses wallet_client
* retry logic via tokio_retry, miner creates new coinbase output via wallet API (retries several times)
* move wallet client into wallet crateand rework the lock acquisition logic to use tokio_retry
* Fix for the chain pipeline partly relying on an outdated head,
leading to not properly recognizing a fork and inconsistent sum
tree state.
* Do not drop block requests during sync that don't get satisfied,
retry enough time to get them and avoid stall.
* Always validate header, even in sync where we may have validated
it already. We don't want a block coming from a peer that could
squeeze through with an invalid header.
* When syncing, do not mark blocks that were errored by the chain
as received (typical case: orphan). Keep retrying.
* Improved chain state dump for debugging.
* Do not add to orphans blocks too far in the future.
* Better error reporting on db errors.
* Related sync test fixes.
TODO figure out why syncing peers timeout so often, very useful
to test but not that great for a fast sync experience.
Addresses a corner case of sync. If we're still in sync mode but
just caught up to the head, a block could be added through normal
gossip. So we can't short-circuit some of the header handling
even though during sync the header should have already been
validated and saved, because we can still get a block from gossip.
* rename pubkey -> key_id, these are key identifiers, not public keys
* reduce identifier seize to 10 bytes, get rid of fingerprints (identifiers are now small enough to use as-is)
* IDENTIFIER_SIZE const
* add FeeDispute error for when sender and recipient disagre on fee calculation (should never happen)
* does not need to be mut
* cleaned up some warnings
* added global slog instance, changed all logging macro formats to include logger instance
* adding configuration to logging, allowing for multiple log outputs
* updates to test, changes to build docs
* rustfmt
* moving logging functions into util crate
* add lock_height to Transaction and TxKernel, sign msg including both fee and lock_height in kernel
* make the order of the fields in tx and kernel more consistent
* rename to kernel_sig_msg
* add test to cover adding timelocked tx to pool, where tx is invalid based on current height of the blockchain
* add tests for adding timelocked txs to blocks (valid and otherwise)
Configuration for a minum accept fee base for the transaction
pool. The base is multipled by a weight computed from the
transaction number of inputs, outputs and kernels. The transaction
fee is required to always be larger than the weight times the
base.
min_fee = base * (-1*input_len + 4*output_len + kernel_len)
The weight is set to never be less than one.
Also added a configurable (and fairly naive for now) max pool
capacity in number of transactions.
* store wallet output data in hashmap
* cleanup up commented out code
* pass pubkey/identifier and not derivation around to miner etc.
* fix failing tests
Update coinbase building and block summation to account for half of
the fees going to the coinbase. Forcing fees to be even as a
consequence. Now that we can't build the coinbase independently
from the block (because fees), had to update the miner to keep the
key derivation so a new derivation isn't made any time a new block
gets worked on.
Minor doc and warning cleanups.
* introduce grin_keychain, encapsulate derivation of secret_keys
* core compiles against keychain, tests don't run yet
* core tests are now passing against keychain
* wip - getting wallet working with keychain
* add util and keychain to travis test matrix
* basic test around key derivation
* Integrate PMMR and its persistent backend with the Chain
* Chain can set tree roots; PMMR backend discard
* Check spent and prune for each input in new block
* Handling of forks by rewinding the state
* More PMMR tests and fixes, mostly around rewind
* Rewrite get_unspent to use the sumtrees, fix remaining compilation issues
* Adding ability to load multiple mining plugins in parallel via cuckoo miner
* updating with newest cuckoo-miner changes
* revert default config value
* update pow document with link to podcast
* workaround the "commitment is not a public key" problem by using to_two_keys() from rust-secp256k1-zkp
* remove secp256k1zkp, update the various Cargo.toml to use rust-secp256k1-zkp from external git repo, update code to not rely on to_pubkey (and use verify_from_commit)
* secp256k1zkp from git
* cleanup secp256k1 dependencies
* gitignore node1 and node2 dirs
* use head_header in add_to_memory_pool
* add COINBASE_MATURITY const to consensus
* add coinbase maturity (wip) validaton rule to validate_block
* add coinbase maturity check to validate_block
* map errors in adapters - specific errors still wip
* reworked so adapter translates chain errors to pool errors (core errors not required)
* add test for spending immature coinbase in memory pool
* wip - add test to cover spending coinbase output in chain.process_block
* added test coverage around process_block - we have a problem with coinbase output commitments
* add a comment on the failing test
* process_block will now fail validation if we attempt to spend coinbase that has not yet matured
(remember to use a new reward_key for every block). test coverage in place to verify this
* Adding output stats from plugin, when available
* adding grin config
* moving pow crate from core/pow to it's own crate
* moving POW/mining stuff into pow module, and removing mining dependency from chain
* refactored most of mining into pow module...miner.rs still in server
* update grin tests
* updating genesis block mining to use configured miner, if available
* chain tests back into chain, done with pow refactor
* use tag of cuckoo-miner for pr
* playing around with changing cuckoo sizes on the fly
* modifying tests to use global cuckoo parameters, and checking results
* check for pow size
* Changing global function names, and removing length from POW serialization
* changes to use blake2 for cuckoo seed generation
* updating tests to blake 2
* minor cleanup - unused imports
* cleanup build warnings - unused vars
* make structs pub to get rid of the private_in_public lint warning
* missing docs on RangeProof
* add missing docs to store delete function
* cleaned up deprecation warning -
tokio_core -> tokio_io
complete() -> send()
* playing around with changing cuckoo sizes on the fly
* modifying tests to use global cuckoo parameters, and checking results
* check for pow size
* Changing global function names, and removing length from POW serialization
* Adding ability to serialise parts of the header, pre-nonce and post-nonce
* Some test integration of queueing functions in cuckoo-miner
* more cuckoo-miner async mode integration, now more or less working
* integrating async miner workflow
* rocksdb update
* u64 internal difficulty representation, and integration of latest Cuckoo-miner API
* change to cuckoo-miner notify function
* Issue in testing, and if use_async value is None in grin.toml
* making async mode explicit in tests - 2
* fiddle with port numbers for CI
* update tag to ensure cuckoo-miner build doesn't fail on windows
* change the order in which tests are run
When a new block is rejected by the validation pipeline as orphan,
the chain now keeps it in a ring buffer of capped size. Anytime
a new block is accepted, all orphans in the ring buffer are
retried. This is helpful to handle all the cases where we receive
blocks out of order for networking/timing reasons.
* Adding cuckoo-miner into grin as a proper dependency now
* Defaulting to using cuckoo-miner
* Updates to tests to use cuckoo_miner by default, (using cuckoo16)
* addition of configuration file to grin executable, instructions on how it's picked up are in grin.toml
* Starting to add a configuration file, and change some elements of configuration structures to make them more consistent
* Refactoring to allow for different miner implementations. Added conditional support for compiling and loading the cuckoo-miner plugin project.
* Small changes to experimentally integrate with cuckoo-miner and compatibility with latest version of cuckoo-miner.
* Turning off inclusion of cuckoo_miner by default
* Disabling simulate_parallel_mining test for now
Introduce a new Chain struct that maintains the current head,
acts as facade to the store and the block processing pipeline
and handles all appropriate locking.
All higher level components, like the overall server, REST APIs
and miner have been update to only depend on Chain. Cleaned up
some duplication and tests now that there's a single entry point
to all blockchain operations.
* Beginning to add a POW description, and some minor changes to mining testing, addition of further debug information to mining output.
* Many additions to create first draft of POW documentation
* Fixes to difficult adjustments by adding a MINIMUM_DIFFICULTY consensus value. Otherwise never adjusted above 1 due to flooring.
* Starting to refactor test, adding http post to wallet sender
* Implemented ability to run servers on different ports (mostly for testing), and implemented ability to post http requests directly to receiving wallets
* Adding detailed instructions on running multiple servers on the same machine
* Changes to build.doc to outline server running process
* Removed unwanted debug statements
* WIP Local server testing framework evolution
* More refactoring of server pool, checked in because there's a problem
* Added server reference structure, and ability to return server references from tests cleanly
* Added simulate_parallel_mining test, which puts some artificial slowdown into test mining loops, and the difficulty can currently be watched in the log
* Removing the ServerRef structure placed in earlier and replaced with a ServerStats structure, that just returns relevant info about the Server state without exposing it to the world
Integrate the new difficulty calculation into the block chain
validation, the miner and tests. As the difficulty calculation
doesn't use varying Cuckoo sizes anymore and we narrowed down
reasonable final Cuckoo Cycle parameters, removed all Cuckoo
Cycle sizes from block headers.
Formalized easier Cuckoo Cycle sizes for testing (and possibly
testnet) by introducing a test mode in configuration. Updated
all tests.
* Beginning to refactor server tests
* Extended test framework to include running wallet servers and miners. Also some changes to the wallet reciever api to include a wallet config and set the data directly for the wallet file explicitly rather than assuming it's always in the working dir.
* Adding wallet config to main executable
* Adding ability to select wallet server while mining
* Configurable wallet receiver url
Fixes a few loose ends in the full cycle of sending coins,
accepting them, pushing that transaction to the pool and having it
mined. More specifically:
* The API output endpoint needs to be a UTXO endpoint, as the
server can't make any guarantee about having a spent output.
* Bubbling up HTTP not found errors.
* Wallet output status checker now handles spent outputs.
* Transaction pool validates the transaction before accepting it.
* Fixed the operation API routes.
* Fixed too greedy wallet coin selection loop.
* Transactions coming from the network are now pushed to the pool
through the net adapter.
* New blocks accepted by the chain are sent to the pool for
eviction.
* The miner requests transactions from the pool to build its
blocks.
* The push API adds to the pool, removing the mock.
* Implementation of the adapter to the chain required by the pool
to get consistent UTXOs. Grossly unoptimized until we have the UTXO
MMR ready.
Most of the logic to build a transaction that sends coin to
another party. Still requires more debugging and clean up.
Main changes and additions are:
* Update to serde 1.0
* API endpoint to retrieve an Output
* Output is now Serialize and Deserialize
* Wallet configuration
* Command line for the send operation
* Wallet data checker to update created outputs into confirmed
* Wallet-specific configuration
With the coinbase receiver daemon in place, when starting a Grin
server in mining mode, the miner will now ask the receiver for a
coinbase output. The output is then used to insert in a block when
successfully mined.
Adds a couple utility function to p2p server to check whether
we're already connected to a peer before trying again. Other
very minor fixes and logging improvements.
