When processing `CMPCTBLOCK` message, at some moments
we can need to process cmpct block txns, since all messages
are handled by ProcessMessage, we call ProcessMessage
all over again. For this reason, it creates a function called
`ProcessCompactBlockTxns` to process it.
DataStream Serialize method has surprising behavior because it just serializes
raw bytes without a length prefix. When you serialize a string or vector, a
length prefix is serialized before the raw object contents so the object can be
unambiguously deserialized later. But DataStreams don't support deserializing
at all and just dump the raw bytes.
Having this inconsistency is not necessary and could be confusing (see
https://github.com/bitcoin/bitcoin/pull/27790#discussion_r1212315030) so this
PR just drops the DataStream::Serialize method.
I commented out a few new lines at the bottom of test/functional/p2p_ibd_txrelay.py
It tests that nodes process certain transactions, after IBD. These new assertions
from upstream are failing, but it may be because our IBD / tx sharing logic is different.
We should revisit this.
fb02a3cd1a p2p: Log addresses of stalling peers (Martin Zumsande)
Pull request description:
This was suggested in #27705 by ArmchairCryptologist.
It allows node operators that have the `-logips` option enabled to better identify potentially misbehaving peers and maybe ban them.
This is especially helpful in case of inbound peers for which (dis)connections aren't logged per default, so it's impossible to use the debug log to connect their `nodeId` to an address unless the very noisy `net` debugging is enabled.
In case of outbound peers for which the address is potentially logged when establishing the connection, this just adds some convenience.
ACKs for top commit:
stratospher:
tACK fb02a3c.
jamesob:
github ACK fb02a3cd1a
0xB10C:
Untested ACK fb02a3cd1a
instagibbs:
utACK fb02a3cd1a
Tree-SHA512: 2080f794c715bd36143405828b4b0e1be859095caf8f8a0c20dd2a4b64d192d78fee0fa350a2bb7c39848718332c4dd4d8edb2cc8d22095b65afe710591f7ccb
A single outbound slot is required, so if the first two slots
are taken by inbound in-flights, the node will reject additional
unless they are coming from outbound.
This means in the case where a fast sybil peer is attempting to
stall out a node, a single high bandwidth outbound peer can
mitigate the attack.
This is a change in behavior so that if for some reason we request a block from a peer, we don't allow an unsolicited CMPCT_BLOCK announcement for that same block to cause a request for a full block from the uninvited peer (as some type of request is already outstanding from the original peer)
5b3406094f net_processing: Boost inv trickle rate (Anthony Towns)
228e9201ef txmempool: have CompareDepthAndScore sort missing txs first (Anthony Towns)
Pull request description:
Couple of performance improvements when draining the inventory-to-send queue:
* drop txs that have already been evicted from the mempool (or included in a block) immediately, rather than at the end of processing
* marginally increase outgoing trickle rate during spikes in tx volume
ACKs for top commit:
willcl-ark:
ACK 5b34060
instagibbs:
ACK https://github.com/bitcoin/bitcoin/pull/27610/commits/5b3406094f2679dfb3763de4414257268565b943
darosior:
utACK 5b3406094f
glozow:
code review ACK 5b3406094f
dergoegge:
utACK 5b3406094f
Tree-SHA512: 155cd3b5d150ba3417c1cd126f2be734497742e85358a19c9d365f4f97c555ff9e846405bbeada13c3575b3713c3a7eb2f780879a828cbbf032ad9a6e5416b30
This is a commit in preparation for the next few commits. The functions
are moved to methods to avoid their re-declaration for the purpose of
passing in BlockManager options.
The functions that were now moved into the BlockManager should no longer
use the params as an argument, but instead use the member variable.
In the moved ReadBlockFromDisk and UndoReadFromDisk, change
the function signature to accept a reference to a CBlockIndex instead of
a raw pointer. The pointer is expected to be non-null, so reflect that
in the type.
To allow for the move of functions to BlockManager methods all call
sites require an instantiated BlockManager, or a callback to one.
If transactions are being added to the mempool at a rate faster than 7tx/s
(INVENTORY_BROADCAST_PER_SECOND) then peers' inventory_to_send queue can
become relatively large. If this happens, increase the number of txids
we include in an INV message (normally capped at 35) by 5 for each 1000
txids in the queue.
This will tend to clear a temporary excess out reasonably quickly; an
excess of 4000 invs to send will be cleared down to 1000 in about 30
minutes, while an excess of 20000 invs would be cleared down to 1000 in
about 60 minutes.
