We should learn about new peers via address messages.
An inbound peer connecting to us tells us nothing about
its ability to accept incoming connections from us, so
we shouldn't assume that we can connect to it based on
this.
The vast majority of nodes on the network do not accept
incoming connections, adding them will only slow down
the process of making a successful connection in the
future.
Nodes which have configured themselves to not announce would prefer we
not violate their privacy by announcing them in GETADDR responses.
Tests if addresses are online or offline by briefly connecting to them. These short lived connections are referred to as feeler connections. Feeler connections are designed to increase the number of fresh online addresses in tried by selecting and connecting to addresses in new. One feeler connection is attempted on average once every two minutes.
This change was suggested as Countermeasure 4 in
Eclipse Attacks on Bitcoin’s Peer-to-Peer Network, Ethan Heilman,
Alison Kendler, Aviv Zohar, Sharon Goldberg. ePrint Archive Report
2015/263. March 2015.
The `pickDataDirectory()` function was calling `exit(0)` to quit
the application when the user closes the dialog without choosing
a data directory.
This is a bad idea because a background thread is created (to
check free space on the drive of the currently selected datadir).
The thread is not stopped and unwound properly, resulting in a potential
race condition somewhere deep in Qt.
So replace the `exit()` by a boolean return value, and let the
stack unwind normally.
qa/rpc-tests/segwit: Test GBT sigops before and after activation
Github-Pull: #8489
Rebased-From: 160f895a80660e4e3904a2624e4110960d051902 239cbd2e5c2a36843b45b356e9aea6e8d35f0968
- Ignore GETBLOCKTXN requests for unknown blocks
Don't disconnect peers, or else we leak information that could be
used for fingerprinting.
- Ignore CMPCTBLOCK messages for pruned blocks
Also ignores CMPCTBLOCK announcements that have too little work. This is to
prevent disk-exhaustion DoS.
Github-Pull: #8408
Rebased-From: 1de2a46632946990a7863020b61172232f8c5796 1d06e49834814eed45e07393dcffd7b6683311b2
de45c06 [Wallet] Add CKeyMetadata record for HDMasterKey(s), factor out HD key generation (Jonas Schnelli)
f142c11 [0.13] Create a new HD seed after encrypting the wallet (Jonas Schnelli)
7945088 [Wallet] comsetic non-code changes for the HD feature (Jonas Schnelli)
68d7682 [Wallet] ensure CKeyMetadata.hdMasterKeyID will be cleared during SetNull() (Jonas Schnelli)
f708085 [QA] extend wallet-hd test to cover HD metadata (Jonas Schnelli)
986c223 [Wallet] print hd masterkeyid in getwalletinfo (Jonas Schnelli)
b1c7b24 [Wallet] report optional HDKeypath/HDMasterKeyId in validateaddress (Jonas Schnelli)
5b95dd2 [Wallet] extend CKeyMetadata with HD keypath (Jonas Schnelli)
since there is i/o that can happen when retrieving extended input
and output information, limiting the duration of the lock in getrawtransaction
can help improve concurrency
There was a previous assumption that blockindex would be quite small. With addressindex
and spentindex enabled the blockindex is much larger and the amount of cache allocated for
it should also increase. Furthermore, enabling compression should decrease the amount of
disk space required and less data to write/read. The default leveldb max_open_files is set to
1000, for the blockindex the default is set to 1000 with compression. The 64 value that is
current is kept for the utxo database and does not enable compression. Two additional options
are added here to be able to configure the values for leveldb and the block index:
- `-dbmaxopenfiles` A number of files for leveldb to keep open
- `-dbcompression` Boolean 0 or 1 to enable snappy leveldb compression
