MAYANode Overview
Overview of MAYANodes
Last updated
Overview of MAYANodes
Last updated
MAYANodes service the MAYAChain network. Each MAYANode is comprised of several independent servers in a cluster. All MAYANodes communicate and operate in cooperation to create a cross-chain swapping network.
Running a node is a serious undertaking. While Node Operators are well compensated for running a node, there are also risks, skills required and costs.
See the Node Operator 101 Video to learn more before running a node.
To set up a node, you have three choices:
Set up manually (not recommended unless you are an expert)
Set up via Kubernetes (recommended)
Set up via Provider (coming soon).
Each MAYANode is comprised of 5 major components.
mayanode - this is a daemon that runs the MAYAChain chain itself and a HTTP server, that gives a RESTful API to the chain.
bifrost - this daemon creates connections to remote chains (like Bitcoin, Ethereum, Binance, etc) to both observe activity on those chains (incoming/outgoing transactions), and also sign/broadcast outgoing transactions (moving funds on remote chains).
gateway: MAYANode gateway proxy to get a single IP address for multiple deployments
midgard - this daemotn is a layer 2 REST API that provides front-end consumers with semi real-time rolled up data and analytics of the MAYAChain network. Most requests to the network will come through Midgard. This daemon is here to keep the chain itself from fielding large quantities of requests. You can think of it as a “read-only slave” to the chain. This keeps the resources of the network focused on processing transactions.
Full nodes - for every chain that is supported by the network, each MAYANode operator will run their own full node of each chain (Bitcoin, Ethereum, Binance, etc).
As new nodes join/leave the network, this triggers a “churning event”. Which means the list of validators that can commit blocks to the chain changes, and also creates a new Asgard vault, while retiring an old one. All funds in this retiring vault are moved to the new Asgard vault.
Normally, a churning event happens every 3 days (50,000 blocks), although it is possible for it to happen more frequently (such as when a node optionally requests to leave the network using the LEAVE memo).
On every churn, the network selects one or more nodes to be churned out of the network (which can be typically churned back in later). In a given churning event, multiple nodes may be selected to be churned out, but never more than 1/3rd of the current validator set. The criterion the network will take into account is the following:
Requests to leave the network (self-removal)
Banned by other nodes (network-removal)
How long an active nodes has been committing blocks (oldest gets removed)
Bad behavior (accrued slash points for poor node operation)
On every churn, the network may select one or more nodes to be churned into the network but never adds more than one to the total. Which nodes that are selected are purely by validator bond size. Larger bond nodes are selected over lower bond nodes.
There is an endpoint on Midgard that has deep analytics in mean and median active & standby bond sizes to drive efficient discovery of the best "bond" size. Maya's minimum bond is currently: 100K Cacao (MIMIR Override).
The network is safe when it is over-bonded, but shrewd Node Operators will probably actively manage their bond and pool part of it instead to maximize yield.
Deciding to run a node should be carefully considered and thought through. While the payoffs/rewards can be significant, there can also be an equally significant cost.
To run a node, you must obtain a significant amount of LP, and minimums apply. This CACAO is sent into the network as a “bond” and held as leverage on each node to ensure they behave in the best interest of the network.
Running a malicious node or stealing from the network results in a slashing of this bond. Here are the ways in which a validator’s bond can get slashed.
Double Sign (5% of minimum bond) - if it is observed that a single validator node is committing blocks on multiple chains. To avoid this, never run two nodes with the same node account at the same time.
Unauthorized transaction (1.5x transaction value) - if a node sends funds without authorization, the bond is slashed 1.5x the value of the stolen funds. The slashed bond is dumped into the pool(s) where the funds were stolen and added to the reserve.
Bond slashing takes directly from the bond and does not affect rewards.
When a node is active, it earns rewards from the network in CACAO. Sufficient rewards are required to be earned for a Validator to be profitable. Running an unreliable node results in rewards being slashed. Here are the ways in which a validator’s rewards can be slashed.
Not Observing (2 slash pts) - if a node does not observe transactions for all chains while other nodes do, they get slash points added.
Not signing a transaction (600 slash pts) - if a node does not sign an outbound transaction, as requested by the network, they will get slash points added.
Fail to keygen (1 hr of revenue) - When the network attempts to churn and attempts to create a new Asgard pubkey for the network and fails to do so due to a specific node(s), they will lose 1 hr of revenue from their bond.
Slash points undo profits made on the network. For every 1 slash point a node receives, they lose 1 block of rewards. Rewards slashing reduces earned rewards and does not affect a validator’s bond.
Node Operators receive rewards if they are bonded and active on the network and are paid out in Cacao. While revenue is generated every block (every 5 seconds) to each operator, those funds are not available to the operator until they churn out of the network. Over time, this incentive increases the median bonded amount, increases the network's security, and allows the network to grow. They're claimed whenever a node leaves the network. See Keeping Track of Rewards below for more details.
Rewards are affected by Incentive Curve. The Incentive Pendulum increases and decreases the amount of CACAO allocated to nodes according to the security and capital leverage of the network.
Dynamic Inflation adds up to rewards. When the system is healthy, there is no inflation; should LPs withdraw CACAO considerably, inflation starts to kick in to incentivize pooled liquidity.
When a node joins the network, the current block height is recorded. The system creates one block unit for every active node for every active block and has a running total of the block units created. When a node leaves, it cashes in its block units for a portion of the bond rewards. The spent block units are destroyed.
For example, there are 10000 CACAO in bond rewards outstanding. Node A has been active for 30 blocks and has 33 block units but accrued 3 slash points. There are 1000 block units in total. Node A leaves the network and cash in its 30 block units (33 - 3). It receives 300 CACAO ((30/1000) * 10000), leaving 9700 CACAO in node rewards. Its 33 block units are destroyed, leaving 967 block units left.
Depending on how the node was set up, it will likely cost between $1000 and $2000 per month, potentially more as the blockchain scales. The main driver of costs is resource allocation to hosting each MAYANode service.
Running a MAYANode is no simple task. As an operator, you must run/maintain multiple Linux servers with extremely high uptime. It is encouraged that only professional systems engineers run nodes to maintain the highest quality, reliability, and security of the network. The simple question to know if you have the skillsets to run a MAYANode is:
Have you used pagerduty before?
If the answer is no, it’s probably best not to run a node and participate in the network in other ways. The following skill sets are required to be an effective node operator.
Advanced knowledge of Linux server administration and security
Advanced knowledge of Kubernetes
Advanced experience running a host of nodes on a hosted platform such as AWS, Google Cloud, Digital Ocean, etc
Knowledge of running full nodes for other chains such as Bitcoin, Ethereum, and Binance.
Willingness to be “on call” at all times to respond to issues when/if your node becomes unavailable
When you run a MAYANode, each MAYANode will have its own node account. An example node account looks like this:
Most importantly, this will tell you how many slash points the node account has accrued, their status, and the size of their bond (which is in 1e8 notation, 1 Cacao == 100000000).
Types of node status:
Unknown - this should never be possible for a valid node account
Whitelisted - node has been bonded, but hasn’t set their keys yet
Standby - waiting to have minimum requirements verified to become “ready” status. This check happens on each churn event (3 days on average).
Ready - node has met minimum requirements to be churned and is ready to do so. Could be selected to churn into the network. Cannot unbond while in this status.
Active - node is an active participant of the network, by securing funds and committing new blocks to the MAYAChain blockchain. Cannot unbond while in this status.
Disabled - node has been disabled by use of LEAVE, and cannot re-join.
To get node account information, make an HTTP call to your maya-api port which will look like the following:
MAYANodes have the ability to vote on Mimir settings.
Mimir settings have specific abilities. The process for voting from a Node is:
Mimir keys and values are listed in the Mimir endpoint.
A node can vote at any time on any key value.
A node's vote is valid as long as they are active (and removed if they are not).
2/3rds of active nodes need to agree for the change to happen
If 2/3rds consensus is not reached, Mimir admin takes priority, or a constant if present.
A node can change its vote anytime.
A node can delete its vote by using -1 value
Voting costs one native transaction fee, which is deducted from their bond.