20 KiB
{% comment %} This file is licensed under the MIT License (MIT) available on http://opensource.org/licenses/MIT. {% endcomment %} {% assign filename="_includes/guide_p2p_network.md" %}
P2P Network
{% include helpers/subhead-links.md %}
{% autocrossref %}
The Bitcoin network protocol allows full nodes ([peers][peer]{:#term-peer}{:.term}) to collaboratively maintain a [peer-to-peer network][network]{:#term-network}{:.term} for block and transaction exchange. Many SPV clients also use this protocol to connect to full nodes.
Consensus rules do not cover networking, so Bitcoin programs may use alternative networks and protocols, such as the [high-speed block relay network][] used by some miners and the [dedicated transaction information servers][electrum server] used by some wallets that provide SPV-level security.
To provide practical examples of the Bitcoin peer-to-peer network, this section uses Bitcoin Core as a representative full node and [BitcoinJ][] as a representative SPV client. Both programs are flexible, so only default behavior is described. Also, for privacy, actual IP addresses in the example output below have been replaced with [RFC5737][] reserved IP addresses.
{% endautocrossref %}
Peer Discovery
{% include helpers/subhead-links.md %}
{% autocrossref %}
When started for the first time, programs don't know the IP
addresses of any active full nodes. In order to discover some IP
addresses, they query one or more DNS names (called [DNS seeds][dns
seed]{:#term-dns-seed}{:.term}) hardcoded into Bitcoin Core and
BitcoinJ. The response to the lookup should include one or more [DNS
A records][] with the IP addresses of full nodes that may accept new
incoming connections. For example, using the [Unix dig
command][dig command]:
;; QUESTION SECTION:
;seed.bitcoin.sipa.be. IN A
;; ANSWER SECTION:
seed.bitcoin.sipa.be. 60 IN A 192.0.2.113
seed.bitcoin.sipa.be. 60 IN A 198.51.100.231
seed.bitcoin.sipa.be. 60 IN A 203.0.113.183
[...]
The DNS seeds are maintained by Bitcoin community members: some of them provide dynamic DNS seed servers which automatically get IP addresses of active nodes by scanning the network; others provide static DNS seeds that are updated manually and are more likely to provide IP addresses for inactive nodes. In either case, nodes are added to the DNS seed if they run on the default Bitcoin ports of 8333 for mainnet or 18333 for testnet.
DNS seed results are not authenticated and a malicious seed operator or network man-in-the-middle attacker can return only IP addresses of nodes controlled by the attacker, isolating a program on the attacker's own network and allowing the attacker to feed it bogus transactions and blocks. For this reason, programs should not rely on DNS seeds exclusively.
Once a program has connected to the network, its peers can begin to send
it addr
(address) messages with the IP addresses and port numbers of
other peers on the network, providing a fully decentralized method of
peer discovery. Bitcoin Core keeps a record of known peers in a
persistent on-disk database which usually allows it to connect directly
to those peers on subsequent startups without having to use DNS seeds.
However, peers often leave the network or change IP addresses, so programs may need to make several different connection attempts at startup before a successful connection is made. This can add a significant delay to the amount of time it takes to connect to the network, forcing a user to wait before sending a transaction or checking the status of payment.
To avoid this possible delay, BitcoinJ always uses dynamic DNS seeds to get IP addresses for nodes believed to be currently active. Bitcoin Core also tries to strike a balance between minimizing delays and avoiding unnecessary DNS seed use: if Bitcoin Core has entries in its peer database, it spends up to 11 seconds attempting to connect to at least one of them before falling back to seeds; if a connection is made within that time, it does not query any seeds.
Both Bitcoin Core and BitcoinJ also include a hardcoded list of IP addresses and port numbers to several dozen nodes which were active around the time that particular version of the software was first released. Bitcoin Core will start attempting to connect to these nodes if none of the DNS seed servers have responded to a query within 60 seconds, providing an automatic fallback option.
As a manual fallback option, Bitcoin Core also provides several
command-line connection options, including the ability to get a list of
peers from a specific node by IP address, or to make a persistent
connection to a specific node by IP address. See the -help text for
details. BitcoinJ can be programmed to do the same thing.
Resources: [Bitcoin Seeder][], the program run by several of the seeds used by Bitcoin Core and BitcoinJ. The Bitcoin Core [DNS Seed Policy][]. The hardcoded list of IP addresses used by Bitcoin Core and BitcoinJ is generated using the [makeseeds script][].
{% endautocrossref %}
Connecting To Peers
{% include helpers/subhead-links.md %}
{% autocrossref %}
Connecting to a peer is done by sending a version message, which
contains your version number, block, and current time to the remote
node. The remote node responds with its own version message. Then both
nodes send a verack message to the other node to indicate the
connection has been established.
Once connected, the client can send to the remote node getaddr and addr messages to gather additional peers.
In order to maintain a connection with a peer, nodes by default will send a message to peers before 30 minutes of inactivity. If 90 minutes pass without a message being received by a peer, the client will assume that connection has closed.
{% endautocrossref %}
Initial Block Download
{% include helpers/subhead-links.md %}
{% autocrossref %}
Before a full node can validate unconfirmed transactions and recently-mined blocks, it must download and validate all blocks from block 1 (the block after the hardcoded genesis block) to the current tip of the best block chain. This is the Initial Block Download (IBD) or initial sync.
Although the word "initial" implies this method is only used once, it can also be used any time a large number of blocks need to be downloaded, such as when a previously-caught-up node has been offline for a long time. In this case, a node can use the IBD method to download all the blocks which were produced since the last time it was online.
Bitcoin Core uses the IBD method any time the last block on its local best block chain has a block header time more than 24 hours in the past. Bitcoin Core 0.10.0 will also perform IBD if its local best block chain is more than 144 blocks lower than its local best headers chain (that is, the local block chain is more than about 24 hours in the past).
{% endautocrossref %}
Blocks-First
{% include helpers/subhead-links.md %}
{% autocrossref %}
Bitcoin Core (up until version [0.9.3][bitcoin core 0.9.3]) uses a simple initial block download (IBD) method we'll call blocks-first. The goal is to download the blocks from the best block chain in sequence.
The first time a node is started, it only has a single block in its
local best block chain---the hardcoded genesis block (block 0). This
node chooses a remote peer, called the sync node, and sends it the
getblocks message illustrated below.
In the header hashes field of the getblocks message, this new node
sends the header hash of the only block it has, the genesis block
(6fe2...0000 in internal byte order). It also sets the stop hash field
to all zeroes to request a maximum-size response.
Upon receipt of the getblocks message, the sync node takes the first
(and only) header hash and searches its local best block chain for a
block with that header hash. It finds that block 0 matches, so it
replies with 500 block inventories (the maximum response to a
getblocks message) starting from block 1. It sends these inventories
in the inv message illustrated below.
Inventories are unique identifiers for information on the network. Each inventory contains a type field and the unique identifier for an instance of the object. For blocks, the unique identifier is a hash of the block's header.
The block inventories appear in the inv message in the same order they
appear in the block chain, so this first inv message contains
inventories for blocks 1 through 501. (For example, the hash of block 1
is 4860...0000 as seen in the illustration above.)
The IBD node uses the received inventories to request 128 blocks from
the sync node in the getdata message illustrated below.
It's important to blocks-first nodes that the blocks be requested and sent in order because each block header references the header hash of the preceding block. That means the IBD node can't fully validate a block until its parent block has been received. Blocks that can't be validated because their parents haven't been received are called orphan blocks; a subsection below describes them in more detail.
Upon receipt of the getdata message, the sync node replies with each
of the blocks requested. Each block is put into serialized block format
and sent in a separate block message. The first block message sent
(for block 1) is illustrated below.
The IBD node downloads each block, validates it, and then requests the
next block it hasn't requested yet, maintaining a queue of up to 128
blocks to download. When it has requested every block for which it has
an inventory, it sends another getblocks message to the sync node
requesting the inventories of up to 500 more blocks. This second
getblocks message contains multiple header hashes as illustrated
below:
Upon receipt of the second getblocks message, the sync node searches
its local best block chain for a block that matches one of the header
hashes in the message, trying each hash in the order they were received.
If it finds a matching hash, it replies with 500 block inventories
starting with the next block from that point. But if there is no
matching hash (besides the stopping hash), it assumes the only block the
two nodes have in common is block 0 and so it sends an inv starting with
block 1 (the same inv message seen several illustrations above).
This repeated search allows the sync node to send useful inventories even if the IBD node's local block chain forked from the sync node's local block chain. This fork detection becomes increasingly useful the closer the IBD node gets to the tip of the block chain.
When the IBD node receives the second inv message, it will request
those blocks using getdata messages. The sync node will respond with
block messages. Then the IBD node will request more inventories with
another getblocks message---and the cycle will repeat until the IBD
node is synced to the tip of the block chain. At that point, the node
will accept blocks sent through the regular block broadcasting described
in a later subsection.
{% endautocrossref %}
Blocks-First Advantages & Disadvantages
{:.no_toc} {% include helpers/subhead-links.md %}
{% autocrossref %}
The primary advantage of blocks-first IBD is its simplicity. The primary disadvantage is that the IBD node relies on a single sync node for all of its downloading. This has several implications:
-
Speed Limits: All requests are made to the sync node, so if the sync node has limited upload bandwidth, the IBD node will have slow download speeds. Note: if the sync node goes offline, Bitcoin Core will continue downloading from another node---but it will still only download from a single sync node at a time.
-
Download Restarts: The sync node can send a non-best (but otherwise valid) block chain to the IBD node. The IBD node won't be able to identify it as non-best until the initial block download nears completion, forcing the IBD node to restart its block chain download over again from a different node. Bitcoin Core ships with several block chain checkpoints at various block heights selected by developers to help an IBD node detect that it is being fed an alternative block chain history---allowing the IBD node to restart its download earlier in the process.
-
Disk Fill Attacks: Closely related to the download restarts, if the sync node sends a non-best (but otherwise valid) block chain, the chain will be stored on disk, wasting space and possibly filling up the disk drive with useless data.
-
High Memory Use: Whether maliciously or by accident, the sync node can send blocks out of order, creating orphan blocks which can't be validated until their parents have been received and validated. Orphan blocks are stored in memory while they await validation, which may lead to high memory use.
All of these problems are addressed in part or in full by the headers-first IBD method used in Bitcoin Core 0.10.0.
Resources: The table below summarizes the messages mentioned throughout this subsection. The links in the message field will take you to the reference page for that message.
| Message | [getblocks][getblocks message] | [inv][inv message] | [getdata][getdata message] | [block][block message]
| From→To | IBD→Sync | Sync→IBD | IBD→Sync | Sync→IBD
| Payload | One or more header hashes | Up to 500 block inventories (unique identifiers) | One or more block inventories | One serialized block
{% endautocrossref %}
Block Broadcasting
{% include helpers/subhead-links.md %}
{% autocrossref %}
At the start of a connection with a peer, both nodes send getblocks messages containing the hash of the latest known block. If a peer believes they have newer blocks or a longer chain, that peer will send an inv message which includes a list of up to 500 hashes of newer blocks, stating that it has the longer chain. The receiving node would then request these blocks using the command getdata, and the remote peer would reply via block messages. After all 500 blocks have been processed, the node can request another set with getblocks, until the node is caught up with the network. Blocks are only accepted when validated by the receiving node.
New blocks are also discovered as miners publish their found blocks, and these messages are propagated in a similar manner. Through previously established connections, an inv message is sent with the new block hashed, and the receiving node requests the block via the getdata message.
{% endautocrossref %}
Orphan Blocks
{% include helpers/subhead-links.md %}
{% autocrossref %}
Blocks-first nodes may download orphan blocks---blocks whose previous block header hash field refers to a block header this node hasn't seen yet. In other words, orphan blocks have no known parent (unlike stale blocks, which have known parents but which aren't part of the best block chain).
When a blocks-first node downloads an orphan block, it will not validate
it. Instead, it will send a getblocks message to the node which sent
the orphan block; the broadcasting node will respond with an inv message
containing inventories of any blocks the downloading node is missing (up
to 500); the downloading node will request those blocks with a getdata
message; and the broadcasting node will send those blocks with a block
message. The downloading node will validate those blocks, and once the
parent of the former orphan block has been validated, it will validate
the former orphan block.
{% endautocrossref %}
Transaction Broadcasting
{% include helpers/subhead-links.md %}
{% autocrossref %}
In order to send a transaction to a peer, an inv message is sent. If a getdata response message is received, the transaction is sent using tx. The peer receiving this transaction also forwards the transaction in the same manner, given that it is a valid transaction.
{% endautocrossref %}
Memory Pool
{% include helpers/subhead-links.md %}
{% autocrossref %}
Full peers may keep track of unconfirmed transactions which are eligible to be included in the next block. This is essential for miners who will actually mine some or all of those transactions, but it's also useful for any peer who wants to keep track of unconfirmed transactions, such as peers serving unconfirmed transaction information to SPV clients.
Because unconfirmed transactions have no permanent status in Bitcoin, Bitcoin Core stores them in non-persistent memory, calling them a memory pool or mempool. When a peer shuts down, its memory pool is lost except for any transactions stored by its wallet. This means that never-mined unconfirmed transactions tend to slowly disappear from the network as peers restart or as they purge some transactions to make room in memory for others.
Transactions which are mined into blocks that later become stale blocks may be added back into the memory pool. These re-added transactions may be re-removed from the pool almost immediately if the replacement blocks include them. This is the case in Bitcoin Core, which removes stale blocks from the chain one by one, starting with the tip (highest block). As each block is removed, its transactions are added back to the memory pool. After all of the stale blocks are removed, the replacement blocks are added to the chain one by one, ending with the new tip. As each block is added, any transactions it confirms are removed from the memory pool.
SPV clients don't have a memory pool for the same reason they don't relay transactions. They can't independently verify that a transaction hasn't yet been included in a block and that it only spends UTXOs, so they can't know which transactions are eligible to be included in the next block.
{% endautocrossref %}
Misbehaving Nodes
{% include helpers/subhead-links.md %}
{% autocrossref %}
Take note that for both types of broadcasting, mechanisms are in place to punish misbehaving peers who take up bandwidth and computing resources by sending false information. If a peer gets a banscore above the -banscore=<n> threshold, he will be banned for the number of seconds defined by -bantime=<n>, which is 86,400 by default (24 hours).
{% endautocrossref %}
Alerts
{% include helpers/subhead-links.md %}
{% autocrossref %}
In case of a bug or attack,
the Bitcoin Core developers provide a
Bitcoin alert service with an RSS feed
and users of Bitcoin Core can check the error field of the getinfo RPC
results to get currently active alerts for their specific version of
Bitcoin Core.
These messages are aggressively broadcast using the alert message, being sent to each peer upon connect for the duration of the alert.
These messages are signed by a specific ECDSA private key that only a small number of developers control.
Resource: More details about the structure of messages and a complete list of message types can be found in the [P2P reference section][section P2P reference].
{% endautocrossref %}