::: warning Disclaimer This is work in progress. Mechanisms and values are susceptible to change. :::
The role of validators is to run a full-node and participate in consensus by broadcasting votes which contain cryptographic signatures signed by their private key. Validators commit new blocks in the blockchain and receive revenue in exchange for their work. They must also participate in governance by voting on proposals. Validators are weighted according to their total stake.
The Xar Hub is a public Proof-Of-Stake (PoS) blockchain, meaning that the weight of validators is determined by the amount of staking tokens (CSDT) bonded as collateral. This CSDT can be self-delegated directly by the validator or delegated to them by other CSDT holders.
Any user in the system can declare their intention to become a validator by sending a
create-validator transaction. From there, they become validator candidates.
The weight (i.e. voting power) of a validator determines whether or not they are an active validator. Initially, only the top 10 validators with the most voting power will be active validators.
A full-node is a program that fully validates transactions and blocks of a blockchain. It is distinct from a light-node that only processes block headers and a small subset of transactions. Running a full-node requires more resources than a light-node but is necessary in order to be a validator. In practice, running a full-node only implies running a non-compromised and up-to-date version of the software with low network latency and without downtime.
Of course, it is possible and encouraged for users to run full-nodes even if they do not plan to be validators.
Delegators are CSDT holders who cannot, or do not want to run a validator themselves. CSDT holders can delegate CSDT to a validator and obtain a part of their revenue in exchange (for more detail on how revenue is distributed, see What is the incentive to stake? and What are validators commission? sections below).
Because they share revenue with their validators, delegators also share risks. Should a validator misbehave, each of their delegators will be partially slashed in proportion to their delegated stake. This is why delegators should perform due diligence on validators before delegating, as well as spreading their stake over multiple validators.
Delegators play a critical role in the system, as they are responsible for choosing validators. Being a delegator is not a passive role: Delegators should actively monitor the actions of their validators and participate in governance.
Any participant in the network can signal that they want to become a validator by sending a
create-validator transaction, where they must fill out the following parameters:
PubKey: The private key associated with this Tendermint
PubKeyis used to sign prevotes and precommits.
Once a validator is created, CSDT holders can delegate CSDT to them, effectively adding stake to their pool. The total stake of an address is the combination of CSDT bonded by delegators and CSDT self-bonded by the entity which designated themselves.
Out of all validator candidates that signaled themselves, the 10 with the most total stake are the ones who are designated as validators. They become validators If a validator’s total stake falls below the top 10 then that validator loses their validator privileges: they don’t participate in consensus and generate rewards any more. Over time, the maximum number of validators will increase, according to the following schedule (Note: this schedule can be changed by governance):
The Testnet is a great environment to test your validator setup before launch.
In short, there are two types of keys:
xarvalconspub(Get this value with
xard tendermint show-validator)
xarcliand used to sign transactions. Application keys are associated with a public key prefixed by
xarpuband an address prefixed by
xar. Both are derived from account keys generated by
xarcli keys add.
Note: A validator’s operator key is directly tied to an application key, but
uses reserved prefixes solely for this purpose:
After a validator is created with a
create-validator transaction, they can be in three states:
in validator set: Validator is in the active set and participates in consensus. Validator is earning rewards and can be slashed for misbehaviour.
jailed: Validator misbehaved and is in jail, i.e. outisde of the validator set. If the jailing is due to being offline for too long, the validator can send an
unjailtransaction in order to re-enter the validator set. If the jailing is due to double signing, the validator cannot unjail.
unbonded: Validator is not in the active set, and therefore not signing blocs. Validator cannot be slashed, and does not earn any reward. It is still possible to delegate CSDT to this validator. Un-delegating from an
unbondedvalidator is immediate.
Self-delegation is delegation from a validator to themselves. This amount can be increases by sending a
delegate transaction from your validator’s
application application key.
The minimum is
Delegators are free to choose validators according to their own subjective criteria. This said, criteria anticipated to be important include:
No, they do not. Each delegator will value validators based on their own criteria. Validators will be able to register a website address when they nominate themselves so that they can advertise their operation as they see fit. Some delegators may prefer a website that clearly displays the team operating the validator and their resume, while others might prefer anonymous validators with positive track records.
Validators have two main responsibilities:
Additionally, validators are expected to be active members of the community. They should always be up-to-date with the current state of the ecosystem so that they can easily adapt to any change.
Validators and delegators on the Xar Hub can vote on proposals to change operational parameters (such as the block gas limit), coordinate upgrades, or make a decision on any given matter.
Validators play a special role in the governance system. Being the pillars of the system, they are required to vote on every proposal. It is especially important since delegators who do not vote will inherit the vote of their validator.
Staking CSDT can be thought of as a safety deposit on validation activities. When a validator or a delegator wants to retrieve part or all of their deposit, they send an
unbonding transaction. Then, CSDT undergo a 3 weeks unbonding period during which they are liable to being slashed for potential misbehaviors committed by the validator before the unbonding process started.
Validators, and by association delegators, receive block rewards, fees, and have the right to participate in governance. If a validator misbehaves, a certain portion of their total stake is slashed. This means that every delegator that bonded CSDT to this validator gets penalized in proportion to their bonded stake. Delegators are therefore incentivized to delegate to validators that they anticipate will function safely.
By delegating to a validator, a user delegates voting power. The more voting power a validator have, the more weight they have in the consensus and governance processes. This does not mean that the validator has custody of their delegators’ CSDT. By no means can a validator run away with its delegator’s funds.
Even though delegated funds cannot be stolen by their validators, delegators are still liable if their validators misbehave.
The validator that is selected to propose the next block is called proposer. Each proposer is selected deterministically, and the frequency of being chosen is proportional to the voting power (i.e. amount of bonded CSDT) of the validator. For example, if the total bonded stake across all validators is 100 CSDT and a validator’s total stake is 10 CSDT, then this validator will proposer ~10% of the blocks.
Yes, they will. If governance decides so, validators of the Xar hub may be required to validate additional zones in the Xar ecosystem.
Each member of a validator’s staking pool earns different types of revenue:
This total revenue is divided among validators’ staking pools according to each validator’s weight. Then, within each validator’s staking pool the revenue is divided among delegators in proportion to each delegator’s stake. A commission on delegators’ revenue is applied by the validator before it is distributed.
Validators earn proportionally more revenue than their delegators because of commissions.
Validators also play a major role in governance. If a delegator does not vote, they inherit the vote from their validator. This gives validators a major responsibility in the ecosystem.
Revenue received by a validator’s pool is split between the validator and their delegators. The validator can apply a commission on the part of the revenue that goes to their delegators. This commission is set as a percentage. Each validator is free to set their initial commission, maximum daily commission change rate and maximum commission. The Xar Hub enforces the parameter that each validator sets. Only the commission rate can change after the validator is created.
Block rewards are distributed proportionally to all validators relative to their voting power. This means that even though each validator gains CSDT with each reward, all validators will maintain equal weight over time.
Let us take an example where we have 10 validators with equal voting power and a commission rate of 1%. Let us also assume that the reward for a block is 1000 CSDT and that each validator has 20% of self-bonded CSDT. These tokens do not go directly to the proposer. Instead, they are evenly spread among validators. So now each validator’s pool has 100 CSDT. These 100 CSDT will be distributed according to each participant’s stake:
100*80%*1% = 0.8 CSDT
100\*20% + Commission = 20.8 CSDT
100\*80% - Commission = 79.2 CSDT
Then, each delegator can claim their part of the 79.2 CSDT in proportion to their stake in the validator’s staking pool.
Fees are similarly distributed with the exception that the block proposer can get a bonus on the fees of the block they propose if they include more than the strict minimum of required precommits.
When a validator is selected to propose the next block, they must include at least 2/3 precommits of the previous block. However, there is an incentive to include more than 2/3 precommits in the form of a bonus. The bonus is linear: it ranges from 1% if the proposer includes 2/3rd precommits (minimum for the block to be valid) to 5% if the proposer includes 100% precommits. Of course the proposer should not wait too long or other validators may timeout and move on to the next proposer. As such, validators have to find a balance between wait-time to get the most signatures and risk of losing out on proposing the next block. This mechanism aims to incentivize non-empty block proposals, better networking between validators as well as to mitigate censorship.
Let’s take a concrete example to illustrate the aforementioned concept. In this example, there are 10 validators with equal stake. Each of them applies a 1% commission rate and has 20% of self-delegated CSDT. Now comes a successful block that collects a total of 1025.51020408 CSDT in fees.
First, a 2% tax is applied. The corresponding CSDT go to the reserve pool. Reserve pool’s funds can be allocated through governance to fund bounties and upgrades.
2% * 1025.51020408 = 20.51020408CSDT go to the reserve pool.
1005 CSDT now remain. Let’s assume that the proposer included 100% of the signatures in its block. It thus obtains the full bonus of 5%.
We have to solve this simple equation to find the reward R for each validator:
9*R + R + R*5% = 1005 ⇔ R = 1005/10.05 = 100
R + R * 5%: 105 CSDT
105 * 80% * 1%= 0.84 CSDT
105 * 20% + Commission= 21.84 CSDT
105 * 80% - Commission= 83.16 CSDT (each delegator will be able to claim its portion of these rewards in proportion to their stake)
100 * 80% * 1%= 0.8 CSDT
100 * 20% + Commission= 20.8 CSDT
100 * 80% - Commission= 79.2 CSDT (each delegator will be able to claim their portion of these rewards in proportion to their stake)
If a validator misbehaves, their delegated stake will be partially slashed. There are currently two faults that can result in slashing of funds for a validator and their delegators:
Yes, they do need to self-delegate at least
1 CSDT. Even though there is no obligation for validators to self-delegate more than
1 CSDT, delegators should want their validator to have more self-delegated CSDT in their staking pool. In other words, validators should have skin in the game.
In order for delegators to have some guarantee about how much skin-in-the-game their validator has, the latter can signal a minimum amount of self-delegated CSDT. If a validator’s self-delegation goes below the limit that it predefined, this validator and all of its delegators will unbond.
For now the community is expected to behave in a smart and self-preserving way. When a mining pool in Bitcoin gets too much mining power the community usually stops contributing to that pool. The Xar Hub will rely on the same effect initially. Other mechanisms are in place to smoothen this process as much as possible:
Validators should expect to provision one or more data center locations with redundant power, networking, firewalls, HSMs and servers.
We expect that a modest level of hardware specifications will be needed initially and that they might rise as network use increases. Participating in the testnet is the best way to learn more.
In addition to running a Xar Hub node, validators should develop monitoring, alerting and management solutions.
The Xar network has the capacity for very high throughput relative to chains like Ethereum or Bitcoin.
We recommend that the data center nodes only connect to trusted full-nodes in the cloud or other validators that know each other socially. This relieves the data center node from the burden of mitigating denial-of-service attacks.
Ultimately, as the network becomes more heavily used, multigigabyte per day bandwidth is very realistic.
A successful validator operation will require the efforts of multiple highly skilled individuals and continuous operational attention. This will be considerably more involved than running a bitcoin miner for instance.
Validators should expect to run an HSM that supports ed25519 keys. Here are potential options:
The Tendermint team does not recommend one solution above the other. The community is encouraged to bolster the effort to improve HSMs and the security of key management.
Running effective operation is the key to avoiding unexpectedly unbonding or being slashed. This includes being able to respond to attacks, outages, as well as to maintain security and isolation in your data center.
Validators should expect to perform regular software updates to accommodate upgrades and bug fixes. There will inevitably be issues with the network early in its bootstrapping phase that will require substantial vigilance.
Denial-of-service attacks occur when an attacker sends a flood of internet traffic to an IP address to prevent the server at the IP address from connecting to the internet.
An attacker scans the network, tries to learn the IP address of various validator nodes and disconnect them from communication by flooding them with traffic.
One recommended way to mitigate these risks is for validators to carefully structure their network topology in a so-called sentry node architecture.
Validator nodes should only connect to full-nodes they trust because they operate them themselves or are run by other validators they know socially. A validator node will typically run in a data center. Most data centers provide direct links the networks of major cloud providers. The validator can use those links to connect to sentry nodes in the cloud. This shifts the burden of denial-of-service from the validator’s node directly to its sentry nodes, and may require new sentry nodes be spun up or activated to mitigate attacks on existing ones.
Sentry nodes can be quickly spun up or change their IP addresses. Because the links to the sentry nodes are in private IP space, an internet based attacked cannot disturb them directly. This will ensure validator block proposals and votes always make it to the rest of the network.
It is expected that good operating procedures on that part of validators will completely mitigate these threats.