@ -7,28 +7,48 @@ Syncing is the process by which Geth catches up to the latest Ethereum block and
## Full nodes
There are three types of full node with different sync modes:
There are two types of full node that use different mechanisms to sync up to the head of the chain:
### Full
### Snap (default)
A snap sync'd node holds the most recent 128 blocks in memory, so transactions in that range are always accessible. However, snap-sync only starts processing from a relatively recent block (as opposed to genesis for a full node). Between the initial sync block and the 128 most recent blocks, the node stores occasional checkpoints that can be used to rebuild the state on-the-fly. This means transactions can be traced back as far as the block that was used for the initial sync. Tracing a single transaction requires reexecuting all preceding transactions in the same block **and** all preceding blocks until the previous stored snapshot. Snap-sync'd nodes are therefore functionally equal to full nodes, but the initial synchronization required a checkpoint block to sync from instead of independently verifying the chain all the way from genesis. Snap sync then only verifies the proof-of-work and ancestor-child block progression and assumes that the state transitions are correct rather than re-executing the transactions in each block to verify the state changes. Snap sync is much faster than full sync. To start a node with snap sync pass `--syncmode snap` at startup.
Snap sync starts by downloading the blocks between the head of the chain and the previous checkpoint. Then it downloads the leaves of the state trie for each block without the intermediate nodes. The state trie is then regenerated locally. The state download is the part of the snap-sync that takes the most time to complete and the progress can be monitored using the ETA values in the log messages. However, the blockchain is also progressing at the same time and invalidating some of the regenerated state data. This means it is also necessary to have a 'healing' phase where errors in the state are fixed. It is not possible to monitor the progress of the state heal because the extent of the errors cannot be known until the current state has already been regenerated. The healing has to outpace the growth of the blockchain, otherwise the node will never catch up to the current state. There are some hardware factors that determine the speed of the state healing (speed of disk read/write and internet connection) and also the total gas used in each block (more gas means more changes to the state that have to be handled).
A full sync generates the current state by executing every block starting from the genesis block. A full sync is the trust-minimized option since there it does not rely upon any information from any trusted source apart from the genesis configuration. All other information is indendently verified as the node re-executes the entire historical sequence of blocks. Only the most recent 128 blocks are stored in a full node - older blocks are pruned periodically and represented as a series of checkpoints from which any previous state can be regenerated on request. 128 blocks is about 25.6 minutes of history with a block time of 12 seconds. To create a full node pass `--syncmode full` at startup.
**Note** Snap sync is the default behaviour, so if the `--syncmode` value is not passed to Geth at startup, Geth will use snap sync. A node that is started using `snap` will switch to block-by-block sync once it has caught up to the head of the chain.
### Archive
### Full
A full sync generates the current state by executing every block starting from the genesis block. A full sync indendently verifies proof-of-work and block provenance as well as all state transitions by re-executing the transactions in the entire historical sequence of blocks. Only the most recent 128 blocks are stored in a full node - older blocks are pruned periodically and represented as a series of checkpoints from which any previous state can be regenerated on request. 128 blocks is about 25.6 minutes of history with a block time of 12 seconds. To create a full node pass `--syncmode full` at startup.
An archive node is a full-sync'd node that retains all historical data right back to genesis. There is no need to regenerate any data from checkpoints because all data is directly available in the node's own storage. Archive nodes are therefore ideal for making fast queries about historical states. At the time of writing (August 2022) an archive node occupies nearly 12 TB of disk space (keep up with the current size on [Etherscan](https://etherscan.io/chartsync/chainarchive)). Archive nodes are created by configuring Geth's garbage collection so that old data is never deleted: `geth --syncmode full --gcmode archive`.
## Archive nodes
### Snap
An archive node is a node that retains all historical data right back to genesis. There is no need to regenerate any data from checkpoints because all data is directly available in the node's own storage. Archive nodes are therefore ideal for making fast queries about historical states. At the time of writing (September 2022) a full archive node that stores all data since genesis occupies nearly 12 TB of disk space (keep up with the current size on [Etherscan](https://etherscan.io/chartsync/chainarchive)). Archive nodes are created by configuring Geth's garbage collection so that old data is never deleted: `geth --syncmode full --gcmode archive`.
A snap sync'd node holds the most recent 128 blocks in memory, so transactions in that range are always accessible. However, snap-sync only starts processing from a relatively recent block (as opposed to genesis for a full node). Between the initial sync block and the 128 most recent blocks, the node stores occasional checkpoints that can be used to rebuild the state on-the-fly. This means transactions can be traced back as far as the block that was used for the initial sync. Tracing a single transaction requires reexecuting all preceding transactions in the same block **and** all preceding blocks until the previous stored snapshot. Snap-sync'd nodes are therefore functionally equal to full nodes, but the initial synchronization required some trusted information (a checkpoint block) to sync from instead of independently verifying the chain all the way from genesis. Snap sync is much faster than full sync. To start a node with snap sync pass `--syncmode snap` at startup.
It is also possible to create a partial/recent archive node where the node was synced using `snap` but the state is never pruned. This creates an archive node that saves all state data from the point that the node first syncs. This is configured by starting Geth with `--syncmode snap gcmode archive`.
## Light nodes
A light node syncs very quickly and stores the bare minimum of blockchain data. Light nodes only process block headers, not entire blocks. This greatly reduces the computation time, storage and bandwidth required relative to a full node. This means light nodes are suitable for resource-constrained devices and can catch up to the head of the chain much faster when they are new or have been offline for a while. The trade-off is that light nodes rely heavily on data served by altruistic full nodes. A light client can be used to query data from Ethereum and submit transactions, acting as a locally-hosted Ethereum wallet. However, because they don't keep local copies of the Ethereum state, light nodes can't validate blocks in the same way as full nodes - they have to trust that the data they receive is honest. To start a node in light mode, pass `--syncmode light`. Be aware that full nodes serving light data are relative scarce so light nodes can struggle to find peers.
A light node syncs very quickly and stores the bare minimum of blockchain data. Light nodes only process block headers, not entire blocks. This greatly reduces the computation time, storage and bandwidth required relative to a full node. Light nodes are suitable for resource-constrained devices and can catch up to the head of the chain much faster when they are new or have been offline for a while. The trade-off is that light nodes rely heavily on data served by altruistic full nodes. A light client can be used to query data from Ethereum and submit transactions, acting as a locally-hosted Ethereum wallet. However, because they don't keep local copies of the Ethereum state, light nodes can't validate blocks in the same way as full nodes - they have to trust that the data they receive is honest. To start a node in light mode, pass `--syncmode light`. Be aware that full nodes serving light data are relative scarce so light nodes can struggle to find peers.
Read more about light nodes on our [LES page](/content/docs/fundamentals/les.md).
{% include note.html content="Light nodes do not currently work on proof-of-stake Ethereum, but they are expected to ship soon!" %}
## Consensus layer syncing
Since Ethereum transitioned to proof-of-stake, all consensus logic and block propagation was handed over to consensus clients. This means that syncing the blockchain is now a process shared between the consensus and execution clients. Blocks are downloaded by the consensus client and verified by the execution client. There are two ways to sync a consensus client: optimistic sync and checkpoint sync.
### Optimistic sync
Optimistic sync downloads blocks before the execution client has validated them. This means the execution client can constantly be fed with up-to-date states to snap-sync to. Assuming the blocks are valid, eventually the sync catches up to the optimistically downloaded head and the blockchain is in sync. From this point, verification is done block-by-block (i.e. the sync mode switches to `full`). In optimistic sync the node assumes the data it receives from its peers is correct during the downloading phase but then retroactively verifies each downloaded block. Nodes are not allowed to attest or propose blocks while they are still 'optimistic' because they can't yet guarantee their view of the ehad of the chain is correct.
Read more in the [optimistic sync specs](https://github.com/ethereum/consensus-specs/blob/dev/sync/optimistic.md).
### Checkpoint sync
Alternatively, the consensus client can grab a checkpoint from a trusted source which provides a target state to sync up to, before switching to full sync and verifying each block in turn. In this mode, the node trusts that the checkpoint is correct. There are many possible sources for this checkpoint - the gold standard would be to get it out-of-band from a trusted friend, but it could also come from block explorers or public APIs/web apps.
**Note** it is not currently possible to use a Geth light node as an execution client on proof-of-stake Ethereum.
## Summary
There are several ways to sync a Geth node. The default is to use snap sync to create a full node. This verifies all blocks starting at a recent checkpoint. A trust-minimized alternative is full-sync, which verifies every block since genesis. These modes prune the blockchain data older than 128 blocks, keeping only checkpoints that enable on-request regeneration of historical states. For rapid queries of historical data an archive node is required. Archive nodes keep local copies of all historical data right back to genesis - currently about 12 TB and growing. The opposite extreme is a light node that doesn't store any blockchain data - it requests everything from full nodes. These configurations are controlled by passing `full`, `snap` or `light` to `--syncmode` at startup. For an archive node, `--syncmode` should be `full` and `--gcmode` should be set to `archive`.
There are several ways to sync a Geth node. The default is to use snap sync to create a full node. This verifies all blocks starting at a recent checkpoint. A trust-minimized alternative is full-sync, which verifies every block since genesis. These modes prune the blockchain data older than 128 blocks, keeping only checkpoints that enable on-request regeneration of historical states. For rapid queries of historical data an archive node is required. Archive nodes keep local copies of all historical data right back to genesis - currently about 12 TB and growing. The opposite extreme is a light node that doesn't store any blockchain data - it requests everything from full nodes. These configurations are controlled by passing `full`, `snap` or `light` to `--syncmode` at startup. For an archive node, `--syncmode` should be `full` and `--gcmode` should be set to `archive`. At the transition to proof-of-stake, light-sync will no longer work (until new light client protocols are shipped).
Joe
2 years ago