Catchup And status
When first starting a node it will process all blocks in the blockchain, even if it does not store all blocks locally. The node does so to verify every block in the blockchain thereby checking the validity of the chain. The process can be time-consuming but is essential when running a trusted node.
With adequate hardware and Internet connection, syncing MainNet takes between 2 and 4 weeks in October 2022.
If you cannot wait for catchup, there are multiple options: * Sandbox allows quick setup of private network and public networks. For public networks, the node will be non-archival and use fast catchup. Sandbox should only be used for development purposes. * Fast Catchup can be used to quickly sync a non-archival node, but requires trust in the entity providing the catchpoint. * Third-party snapshots such AlgoNode.io snapshots may be used for archival nodes, but requires trust in the third party. Algorand denies any responsibility if any such snapshot is used. * Use of a third-party API service service.
If a node is stopped it will stop processing blocks. Once the node is restarted, it will start processing blocks where it left off.
The terms "sync" and "catchup/catch up" are used interchangeably.
It is possible to check the status of the catchup process by checking a node's status.
goal node status [-d <data directory>]
After running this status check, monitor the
Sync Time: property that is returned. If this value is incrementing, the node is still synching. The
Sync Time: will display
Sync Time: 0.0s when the node is fully caught up. The status also reports the last block process by the node in the
Last committed block: property. Comparing this block number to what is shown using an Algorand Block Explorer will indicate how much more time catchup will take.
Progress of last committed block is not linear. Earlier blocks usually have much fewer transactions than more recent blocks and catchup usually slows down the closer to the current block you are.