State Syncing [Desmos]

Hello Cat Boss here! Today we’re going to talk about state syncing.

Whether you have run out of space or need to quickly sync up a node, this is a skillset we believe validators should master

The original documentation can be found here: https://docs.desmos.network/mainnet/state-sync

Note: This method works with most Cosmos SDK chains but we will be focusing on Desmos in this post. (And we say this with caution, please test this out on a new server and make sure it works before doing anything you’ll regret)

Prerequisites

Before we start you’ll need access to a RPC that has snapshots enabled. If it is your own you can check it at `/.desmos/data/snapshots`

If you’re unable to get snapshots you can always set it up on a node inside `/.desmos/config/app.toml`. Make sure to restart the node if you make any changes to this file.

###############################################################################
###                        State Sync Configuration                         ###
###############################################################################
# State sync snapshots allow other nodes to rapidly join the network without replaying historical
# blocks, instead downloading and applying a snapshot of the application state at a given height.
[state-sync]
# snapshot-interval specifies the block interval at which local state sync snapshots are
# taken (0 to disable). Must be a multiple of pruning-keep-every.
snapshot-interval = 1500
# snapshot-keep-recent specifies the number of recent snapshots to keep and serve (0 to keep all).
snapshot-keep-recent = 2

Proceed to the next steps once you start having snapshots.

Setup

In a new node setup we should enable state sync inside `/.desmos/config/config.toml`.

Below you will want to fill out the following:

rpc_servers — This is where we put in the rpc with snapshots. While state sync requires 2 rpcs, we have found out they don’t stop you from using the same RPC twice. So if this is a RPC you fully trust we believe it’s okay to just repeat it like the example below.

trust_height — get the latest block height and go a ~thousand blocks back. ex — if the latest block height is 1415137 we would use 1414000. There is no right or wrong here but we have found going a thousand blocks back works better than going a few hundred.

trust_hash — the block hash of your trust_height

#######################################################
###         State Sync Configuration Options        ###
#######################################################
[statesync]
# State sync rapidly bootstraps a new node by discovering, fetching, and restoring a state machine
# snapshot from peers instead of fetching and replaying historical blocks. Requires some peers in
# the network to take and serve state machine snapshots. State sync is not attempted if the node
# has any local state (LastBlockHeight > 0). The node will have a truncated block history,
# starting from the height of the snapshot.
enable = false
# RPC servers (comma-separated) for light client verification of the synced state machine and
# retrieval of state data for node bootstrapping. Also needs a trusted height and corresponding
# header hash obtained from a trusted source, and a period during which validators can be trusted.
#
# For Cosmos SDK-based chains, trust_period should usually be about 2/3 of the unbonding time (~2
# weeks) during which they can be financially punished (slashed) for misbehavior.
rpc_servers = "<rpc>,<rpc>"
trust_height = 1414000
trust_hash = "81FBAC1952121951E7A3D485CC8908D5BA30D3C4E14738F6FFA1C59A0E9F6886"
trust_period = "336h0m0s"

Test Run

If this is not a new server then please run desmos unsafe-reset-all to remove all existing data (make sure you understand what this does and backup anything you may need).

We can now test this out by running desmos start . If it is successful you should see something like the following:

Then finally you should see something like this,

12:28PM INF Snapshot restored format=1 hash="uu��7��[G�5p����\x16��*^|U���V�iʵw" height=1400000 module=statesync

Followed by a bunch of commits to catchup from the snapshot to the latest block height:

Once it’s okay, feel free to stop it and run it back as a background service.

If you check the server size now you should see how empty and happy it is
ヾ(=`ω´=)ノ”)