Substrate and Polkadot Stack Exchange is a question and answer site for developers building blockchains with the Substrate SDK. It only takes a minute to sign up.
Sign up to join this community
It's my understanding that the relay chain runs these types of node types (configurations):
validators, nominators, and archive nodes. fishermen have been deprecated/removed.
Does it also run full, rpc, or boot? Anything else?
For parachains, they run:
collators, full, archive, rpc, boot
Anything else?
I think your mental model is a little off here, at least the way it is written.
There is a binary specific for the Polkadot relay chain node, and software specific for each parachain node. Each of these node binaries have a set of configurations on how you want to run that node, and a node can play multiple of those roles.
For example:
archive node: this is a configuration where a node will keep all blocks and the full state of any block in the database. Lots of nodes do not need all this information, so these choose to prune their blocks, saving disk space. You do not need to be an archive node to participate in the network, but you will need it if you want to easily query data from the past.
full node: this is basically a non-archive node, but can still participate in block production. Usually this means keeping enough blocks around to allow for block reorganization before things are finalized.
block producing nodes: there are nodes which are configured to produce blocks. For the relay chain these are validators, and for parachains these are collators. Their specific role is different in the overall Polkadot protocol, but at an abstract level, you can think of them both as block producers.
boot nodes: these nodes which are not configured in a specific way, but are just nodes which have a well known network address, so that new nodes can start to peer with others on the network. Any node above could play this role.
rpc nodes: these are nodes which publicly expose their RPC endpoints. Any of the nodes above could also be an RPC node, but usually it requires a little bit of extra security considerations, since exposing your RPC nodes also exposes you to people who want to DoS spam your node with requests.
So as you can see, you can have a single node which is an archive, boot, and rpc node based on its configuration.
The one thing you definitely got wrong is "nominator". There is no nominator node. As a nominator, you do not need to run any software, you just need to configure your nominations on the blockchain, and everything is handled by the network.
