Deep dive

DHT stands for Distributed Hash Table, which is a type of distributed key-value database. In this system, each member of the network can store information, such as details about themselves.

The implementation of DHT in TON is similar to the Kademlia protocol, which is also used in IPFS.

Any network participant can operate a DHT node, generate keys, and store data. To do this, they need to create a random ID and inform other nodes about their presence.

An algorithm determines the "distance" between the node and the key, which helps identify which node should store the data. The algorithm is straightforward: it takes the node's ID and the key's ID and performs the XOR operation. A smaller resulting value indicates a closer proximity between the node and the key.

The goal is to store the key on nodes that are as close as possible to the key so that other network participants can, using the same algorithm, easily locate a node that can provide data associated with that key.

Finding a value by key

Let's examine an example involving a search for a key: connect to any DHT node and establish a connection via ADNL UDP.

Suppose we want to find the address and public key needed to connect to the node hosting the foundation.ton site.

Assuming we have already obtained this site's ADNL address by executing the "get method" of the DNS contract, the ADNL address in hexadecimal format is 516618cf6cbe9004f6883e742c9a2e3ca53ed02e3e36f4cef62a98ee1e449174.

Our objective is to determine the IP address, port number, and public key of the node associated with this address.

To achieve this, we first need to get the ID of the DHT key. We will begin by populating the DHT key schema:

dht.key id:int256 name:bytes idx:int = dht.Key

The term name refers to the type of key. For ADNL addresses, the term address is used. For instance, when searching for shard nodes, the term nodes is used. However, the key type can vary and may consist of any array of bytes, depending on the specific value you are seeking.

By applying this schema, we get:

8fde67f6                                                           -- TL ID dht.key
516618cf6cbe9004f6883e742c9a2e3ca53ed02e3e36f4cef62a98ee1e449174   -- the ADNL address being searched
07 61646472657373                                                  -- key type, the word "address" as a TL array of bytes
00000000                                                           -- index 0 because there is only 1 key

Next, retrieve the key ID and the SHA-256 hash from the bytes serialized above. It will be b30af0538916421b46df4ce580bf3a29316831e0c3323a7f156df0236c5b2f75.

Now we can begin our search. To do this, we need to execute a query that uses the following schema:

dht.findValue key:int256 k:int = dht.ValueResult

The key represents the ID of our DHT key, while k indicates the "width" of the search. A smaller value for k results in a more accurate search but limits the number of potential nodes to query. In TON, the specific value of k depends on the implementation.

Now, let's populate this structure, serialize it, and send the request using the adnl.message.query schema. For more details, please refer to the documentation here.

In response, we can get:

• dht.valueNotFound - if the value is not found.
• dht.valueFound - if the value is found on this node.

dht.valueNotFound

If we receive dht.valueNotFound, the response will include a list of nodes that are known to the node we queried and as close as possible to the key we requested. In this situation, we need to connect to these received nodes and add them to our list of known nodes.

Afterwards, we will select the closest, accessible nodes that have not yet been queried from our entire list of known nodes and send the same request to one of them. We will continue this process until we have tried all the nodes within our chosen range or until we stop receiving new nodes.

Let’s analyze the response fields and the schemas used in more detail:

adnl.address.udp ip:int port:int = adnl.Address;
adnl.addressList addrs:(vector adnl.Address) version:int reinit_date:int priority:int expire_at:int = adnl.AddressList;

dht.node id:PublicKey addr_list:adnl.addressList version:int signature:bytes = dht.Node;
dht.nodes nodes:(vector dht.node) = dht.Nodes;

dht.valueNotFound nodes:dht.nodes = dht.ValueResult;

dht.nodes -> nodes - list of DHT nodes (array).

Each node has an id, which serves as its public key, typically represented as pub.ed25519. This key is used to connect to the node via ADNL. Additionally, each node contains a list of addresses, addr_list:adnl.addressList, along with its version and signature.

We need to verify the signature of each node. To do this, we first read the value of the signature field and then set it to zero, effectively making it an empty byte array. Next, we serialize the TL structure dht.node using this empty signature and check the signature field that we emptied earlier.

We validate the serialized bytes using the public key from the id field. [Please see implementation example].

From the list addrs:(vector adnl.Address), we select an address and attempt to establish an ADNL UDP connection, using id (the public key) as the server key.

To determine the "distance" to this node, we retrieve the key ID from the id field and calculate the distance using the XOR operation between the node's key ID and the desired key. If the distance is small enough, we can make the same request to this node. This process continues until we find a value or run out of new nodes.

dht.valueFound

The response will include the value itself, complete key information, and optionally a signature, depending on the value type.

Let's analyze the response fields in more detail, the schemas used:

adnl.address.udp ip:int port:int = adnl.Address;
adnl.addressList addrs:(vector adnl.Address) version:int reinit_date:int priority:int expire_at:int = adnl.AddressList;

dht.key id:int256 name:bytes idx:int = dht.Key;

dht.updateRule.signature = dht.UpdateRule;
dht.updateRule.anybody = dht.UpdateRule;
dht.updateRule.overlayNodes = dht.UpdateRule;

dht.keyDescription key:dht.key id:PublicKey update_rule:dht.UpdateRule signature:bytes = dht.KeyDescription;

dht.value key:dht.keyDescription value:bytes ttl:int signature:bytes = dht.Value;

dht.valueFound value:dht.Value = dht.ValueResult;

Let's determine key:dht.keyDescription. This provides a complete description of the key, including the key itself and information about who can update its value and how.

• key:dht.key - the key must match the one from which we took the key ID for the search.
• id:PublicKey - the public key of the record owner.
• update_rule:dht.UpdateRule - record update rule.
  • dht.updateRule.signature - only the owner of the private key can update the record, the signature of both the key and the value must be valid
  • dht.updateRule.anybody - everyone can update the record, signature is empty and not checked
  • dht.updateRule.overlayNodes - nodes from the same overlay can update the key, used to find nodes of the same overlay and add yourself

dht.updateRule.signature

After reviewing the key's description, we proceed based on the updateRule. In the ADNL address lookup, the type is always dht.updateRule.signature.

We verify the key signature in the same manner as before. First, we set the signature to an empty byte array, serialize it, and perform the necessary checks. Next, we repeat this process for the entire dht.value object while ensuring that the key signature is restored to its original state.

[Please see implementation example].

dht.updateRule.overlayNodes

Used for keys that contain information about other nodes, shards of the WorkChain in the network, the value always follows the TL structure overlay.nodes.

The signature field must be empty.

overlay.node id:PublicKey overlay:int256 version:int signature:bytes = overlay.Node;
overlay.nodes nodes:(vector overlay.node) = overlay.Nodes;

To check for validity, we must check all nodes and for each check signature against its id by serializing the TL structure:

overlay.node.toSign id:adnl.id.short overlay:int256 version:int = overlay.node.ToSign;

We should replace the id with adnl.id.short, which is the key identifier (hash) from the original structure's id field. After serialization, we will verify the signature against the data.

As a result, we obtain a valid list of nodes that can provide information about the required WorkChain shard.

dht.updateRule.anybody

There are no signatures required; anyone can make updates.

Using a value

Once everything has been verified and the ttl:int value has not expired, we can begin working with the value itself, specifically value:bytes. For an ADNL address, this will include an adnl.addressList structure.

This structure will contain the IP addresses and ports of the servers corresponding to the requested ADNL address. In our case, we will have an RLDP-HTTP address associated with the foundation.ton service.

We will use the public key, id:PublicKey, from the DHT key information as the server key.

After establishing the connection, we can request the site's pages using the RLDP protocol. At this stage, the task from the DHT perspective is complete.

Search for nodes that store the state of the blockchain

DHT is also used to locate information about the nodes storing the data of WorkChains and their shards. The process for retrieving this information is similar to searching for any key; however, the key serialization and response validation differ. We will examine these aspects in this section.

To retrieve data, such as that of the MasterChain and its shards, we need to complete the TL structure:

tonNode.shardPublicOverlayId workchain:int shard:long zero_state_file_hash:int256 = tonNode.ShardPublicOverlayId;

In the context of a MasterChain, the workchain value will be set to -1. The corresponding shard will be represented as -9223372036854775808 (0x8000000000000000). Additionally, the zero_state_file_hash refers to the hash of the chain’s zero state (file_hash). Like other data, this can be obtained from the global network configuration in the validator field.

"zero_state": {
  "workchain": -1,
  "shard": -9223372036854775808,
  "seqno": 0,
  "root_hash": "F6OpKZKqvqeFp6CQmFomXNMfMj2EnaUSOXN+Mh+wVWk=",
  "file_hash": "XplPz01CXAps5qeSWUtxcyBfdAo5zVb1N979KLSKD24="
}

Once we fill in tonNode.shardPublicOverlayId, we will serialize it and obtain the key ID by hashing, as we normally do.

Next, we use this key ID as the name to populate the pub.overlay name:bytes = PublicKey structure, wrapping it in a TL bytes array. After serialization, we will retrieve the key ID again from this structure.

This resulting ID will serve as the key for the command:

dht.findValue

When constructing the corresponding dht.key, set name to nodes. We will repeat the process from the previous section; everything remains the same as before, but this time the updateRule will be set to dht.updateRule.overlayNodes.

After the validation process, we will obtain the public keys (IDs) of the nodes that have information about our workchain and shard. To access the ADNL addresses of these nodes, we will hash the keys to create IDs and repeat the same procedure for each ADNL address, similar to how we did for the foundation.ton domain.

As a result, we will have the addresses of the nodes. If desired, we can use these addresses to discover additional nodes within the same chain using overlay.getRandomPeers.

These nodes will also provide us with all the information regarding the blocks.

References

Here is the link to the original article - Oleg Baranov.

Was this article useful?