* Improve handling of RocksDB corruption (#7630)
* kvdb-rocksdb: update rust-rocksdb version
* kvdb-rocksdb: mark corruptions and attempt repair on db open
* kvdb-rocksdb: better corruption detection on open
* kvdb-rocksdb: add corruption_file_name const
* kvdb-rocksdb: rename mark_corruption to check_for_corruption
* Hardening of CSP (#7621)
* Fixed delegatecall's from/to (#7568)
* Fixed delegatecall's from/to, closes#7166
* added tests for delegatecall traces, #7167
* Light client RPCs (#7603)
* Implement registrar.
* Implement eth_getCode
* Don't wait for providers.
* Don't wait for providers.
* Fix linting and wasm tests.
* Problem: AttachedProtocols don't get registered (#7610)
I was investigating issues I am having with Whisper support. I've
enabled Whisper on a custom test network and inserted traces into
Whisper handler implementation (Network<T> and NetworkProtocolHandler
for Network<T>) and I noticed that the handler was never invoked.
After further research on this matter, I found out that
AttachedProtocol's register function does nothing:
https://github.com/paritytech/parity/blob/master/sync/src/api.rs#L172
but there was an implementation originally:
99075ad#diff-5212acb6bcea60e9804ba7b50f6fe6ec and it did the actual
expected logic of registering the protocol in the NetworkService.
However, as of 16d84f8#diff-5212acb6bcea60e9804ba7b50f6fe6ec ("finished
removing ipc") this implementation is gone and only the no-op function
is left.
Which leads me to a conclusion that in fact Whisper's handler never gets
registered in the service and therefore two nodes won't communicate
using it.
Solution: Resurrect original non-empty `AttachedProtocols.register`
implementation
Resolves#7566
* Fix Temporarily Invalid blocks handling (#7613)
* Handle temporarily invalid blocks in sync.
* Fix tests.
* kvdb-rocksdb: update to RocksDB 5.8.8
* kvdb-rocksdb: tune RocksDB options
* Switch to level-style compaction
* Increase default block size (16K), and use bigger blocks for HDDs (64K)
* Increase default file size base (64MB SSDs, 256MB HDDs)
* Create a single block cache shared across all column families
* Tune compaction settings using RocksDB helper functions, taking into account
memory budget spread across all columns
* Configure backgrounds jobs based on the number of CPUs
* Set some default recommended settings
* ethcore: remove unused config blockchain.db_cache_size
* parity: increase default value for db_cache_size
* kvdb-rocksdb: enable compression on all levels
* kvdb-rocksdb: set global db_write_bufer_size
* kvdb-rocksdb: reduce db_write_bufer_size to force earlier flushing
* kvdb-rocksdb: use master branch for rust-rocksdb dependency
The error is "PoW too low to compete with other messages"
This has been previously reported in #7144
Solution: prevent the move semantics
The source of the error is in PoolHandle.relay
implementation for NetPoolHandle.
Because of the move semantics, `res` variable is in fact
copied (as it implements Copy) into the closure and for
that reason, the returned result is always `false.
* created the dir crate in util
* moved code from ethstore/src/dir/paths.rs to dir crate
* rename dir module in ethstore to accounts_dir to distinguish it
from the dir crate
* changes after @tomusdrw on #6952
* parity-version module split from util
removed unused util deps and features
trigger buildbot again
only kvdb links rocksdb
snappy linker issues
* rm snappy
* fixed old version imports
