Nomadic Labs
Nomadic Labs
How to write a Tezos protocol - part 2

This is the second post of a tutorial series on how to implement a Tezos protocol.

In the first post, we saw how to write, compile, register, activate and use an extremely simple protocol. We also looked at the interface between the protocol and the shell.

In this post, we consider a new protocol called demo_counter which extends demo_noops from the first post in several ways.

  • Blocks can contain simple operations, whose effects update the blockchain state.
  • It is parameterized by protocol parameters passed at activation time.
  • It defines REST services (a.k.a. RPCs), in addition to the generic ones already available from the shell.
  • It defines a client library, extending tezos-client with protocol-specific commands.

A large part of this post is devoted to the client library. While this library is not part of the protocol per se, it is needed if we want to communicate with the node in any meaningful way.

This protocol and the client library also make use of additional libraries, such as command-line parsing tools, error monads, RPCs… Describing them in detail would be beyond the scope of this post, but we try to provide enough information to keep the post self-contained.

demo_counter can be found in revision 254be33 of the master branch on the Tezos repository. demo_counter is located in src/proto_demo_counter/.

We refer to the first post for compilation instructions. In most cases, it should be enough to run

make build-deps


The protocol is referred to by the hash ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT. We can check it is indeed known by the node.

# tezos-admin-client list protocols

It is defined by several modules:

ls src/proto_demo_counter/lib_protocol/*.mli
lib_protocol/apply.mli           lib_protocol/main.mli            lib_protocol/receipt.mli
lib_protocol/error.mli           lib_protocol/proto_operation.mli lib_protocol/services.mli
lib_protocol/header.mli          lib_protocol/proto_params.mli    lib_protocol/state.mli

Most protocol-specific types required in Main are now defined in separate modules.

  • block_header_data is defined as Header.t,
  • operation_receipt is defined as Receipt.t,
  • operation_data is defined as Proto_operation.t.

As for demo_noops, block_header_data is still a string, and fitness is defined as the height of the chain. More interesting are the protocol operations and the operation receipt.

The protocol defines three operations in Proto_operation.t, which act on a state State.t stored in the protocol context. As seen in the first post, in the Tezos model each operation is applied to a context and can produce a new context, the context is a map that can be seen as the current state of the blockchain. For proto_counter, the context maps a key "state" to a serialized form of State.t.

State is simply a couple of nonnegative counters (which we can also view as the balances of two accounts).

type t = { a : int32; b : int32 }

Operations are defined in Proto_operation as

type t =
  | IncrA
  | IncrB
  | Transfer of Int32.t (* transfer from A to B, possibly a negative amount *)

The module Apply defines a function

val apply : State.t -> Proto_operation.t -> State.t option

that applies the operation as expected. Some operations may be invalid (and in this case apply returns None). For instance, transfer requires that both counters stay nonnegative, and increment operations require that counters don’t overflow.

Operation application is defined by the function Main.apply_operation.

val apply_operation : validation_state -> operation ->
  (validation_state * operation_receipt) tzresult Lwt.t

let apply_operation validation_state operation =
  Logging.log_notice "apply_operation";
  let { context ; fitness } = validation_state in
  State.get_state context >>= fun state ->
  match Apply.apply state operation.protocol_data with
  | None -> Error.Invalid_operation
  | Some state ->
      let receipt = Receipt.create "operation applied successfully" in
      State.update_state context state >>= fun context ->
      return ({ context ; fitness }, receipt)

This is quite straightforward. If the application succeeds, fitness is left unchanged and the resulting context contains the updated state. This function also returns a receipt that describes the effect of the operation. In this protocol, the receipt is simply a string, but it could be more descriptive. If the application fails, an error is returned via an error monad. All protocol errors are registered in Error.

Protocol parameters

We saw in the first post that when a protocol is activated, we can pass to it initialization parameters through a JSON value. This value is provided by a user through a file argument, e.g., protocol_parameters.json, to the activation command. The demo_noops protocol did not take advantage of this feature, but demo_counter uses a JSON value of the form {'init_a': A, 'init_b': B}, where A and B are the initial values of the counters.

The type of the protocol parameters and their encoding are defined in Proto_params.

By convention, the protocol parameters are stored in the context under the key "protocol_parameters". The activation operation of the genesis protocol sets the parameters under this key, and they are retrieved by demo_counter in Main.init.

RPC Services

The protocol implements two services.

  • /chains/main/blocks/head/counter/a returns the value of counter a
  • /chains/main/blocks/head/counter/b returns the value of counter b

Services rely on the RPC_* modules accessible through the protocol environment. Ultimately, these modules are implemented by the tezos-rpc library.

Services are registered by the function Services.rpc_services, which is called by Main.rpc_services at protocol activation.

val rpc_services : Updater.rpc_context RPC_directory.t

let rpc_services = Services.rpc_services


Two libraries are compiled from the protocol code.

  • tezos-protocol-demo-counter is linked to the client library (see below),
  • tezos-embedded-protocol-demo-counter is linked to the node (see src/bin_node/dune).

Recall from the first post that the protocol hash and modules are given in TEZOS_PROTOCOL.

> cat src/proto_demo_counter/lib_protocol/TEZOS_PROTOCOL
    "hash": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
    "modules": ["Error", "Proto_params", "Header", "State",
                "Proto_operation", "Receipt", "Apply", "Services", "Main"]

The compilation steps are given in a generic dune file common to all protocols, and a protocol-specific file that can be generated from the TEZOS_PROTOCOL file. We will discuss compilation at greater length in the next blog post.

Client library

The client and the node interact using RPCs. In theory, we could write a client for demo_counter from scratch in any language, but it is convenient to simply extend tezos-client with a protocol-specific client library. Hence, we can keep using generic features of tezos-client, such as wallet management, and simply add new commands specific to the new protocol. Moreover, we can use well-tested OCaml libraries to conveniently call shell RPCs, and we can access to the protocol code from the client, for instance to access some of its datatypes or services.

For demo_counter, the client library is tezos-client-demo-counter and it is defined in src/proto_demo_counter/lib_client/. It is linked to tezos-client (see bin_client/dune). It is composed of four modules Client_proto_args, Client_proto_commands, Client_proto_main, Protocol_client_context which we will describe as we go.

We can check that demo_counter is indeed known to the client.

# tezos-admin-client list understood protocols

Note that Proto_demo_noops isn’t in the list since it doesn’t have a client library.

User interface

demo_counter adds a few new commands.

# tezos-client -p ProtoDemoCou man
Commands for protocol Demo_counter:
  bake <message>
    Bake a block
    <message>: message in block header
    Increment A
    Increment B
  transfer <amount>
    transfer from A to B
    <amount>: amount taken from A and given to B (possibly negative)
    Get A counter
    Get B counter

Client/Node interaction

Typically, the client library interacts with the node in two ways.

  • using shell services (through the tezos-shell-services library),
  • using protocol services (through the protocol library, i.e. tezos-protocol-demo-counter).

Let see how to use these libraries in practice.

Library tezos-protocol-demo-counter

The library tezos-protocol-demo-counter contains the protocol code and its environment. The modules of the protocol are grouped in a module Protocol. The environment is accessible from module Protocol.Environment.

Typically, we want to use the datatypes defined in the protocol. For instance, to build blocks, we need access to the type of operations and block header data. We also want to access protocol-defined services. In our case, we can get the counter values using the client stub Services.get_counter.

Remark that although it is feasible, the client code should not use protocol functions who read or write the protocol context.

Library tezos-shell-services

This library defines client stubs to call shell RPC services. The demo_counter client library uses injection services and block services.

As an example, consider the (slightly simplified) function Shell_services.Injection.block defined in module Injection_services in lib_shell_services/.

val block:
  #RPC_context.simple -> MBytes.t -> Operation.t list list
  -> Block_hash.t tzresult Lwt.t
  • RPC_context.simple is the RPC context, whic identifies the server, deals with the networking aspects of the call, and serializes the transmitted values.
  • Mbytes.t contains the encoded block header.
  • Operation.t is the generic shell operation type.
  • Block_hash.t is the hash of the injected block.

This function calls the service /injection/block and serializes the parameters and the returned value as expected.

The client library uses two injection services.

  • Shell_services.Injection.block to call /injection/block,
  • Shell_services.Injection.operation to call /injection/operation.

The other family of services, block services, is defined as a functor parameterized by a protocol type Block_services.PROTO defined in lib_shell_services. This allows the stubs to deserialize the JSON values returned by the services, and to return the actual protocol types to the client.

In, we instantiate the functor Block_services.Make with the Protocol module (there are two occurrences because we need to provide a module for both the current and the next protocol).

module Demo_block_services = Block_services.Make(Protocol)(Protocol)

The client library uses the following stubs:

  • Demo_block_services.hash to call /chains/main/blocks/head/hash
  • Demo_block_services.Mempool.pending_operations to call /chains/main/mempool/pending_operations
  • Demo_block_services.Helpers.Preapply.block to call /chains/main/blocks/head/helpers/preapply/block
  • Demo_block_services.Helpers.Preapply.operations to call /chains/main/blocks/head/helpers/preapply/operations/

Commands implementation

Commands are implemented with the help of several libraries.

  • tezos-clic is a command-line parsing library.
  • tezos-client-commands provides the registration function for new commands.
  • client-base defines notably the Client_context.full class, which contains the client context (e.g. wallet, printing facilities, RPC context…).

The commands’ syntax is defined in the modules Client_proto_main and Client_proto_args. Commands are registered using the function Client_command.register.

Commands behavior is implemented in module Client_proto_commands. A command implementation may use a value of type Client_context.full, which is provided by the registration function. Through this object, commands can access the client and RPC contexts.

More precisely, our command implementations use a Protocol_client_context.full object, which is a specialized version of Client_context.full, defined in Protocol_client_context.

Let us have a closer look to the block baking command Client_proto_commands.bake.

let bake (cctxt : Protocol_client_context.full) message : unit tzresult Lwt.t =
  Demo_block_services.Mempool.pending_operations cctxt ()
  >>=? fun { applied; _} ->
  let operations = snd applied in
  let block_header_data = Header.create message in
  Demo_block_services.Helpers.Preapply.block cctxt [operations] ~protocol_data:block_header_data
  >>=? fun (shell, preapply_result) ->
  let block_header_data_encoded =
    Data_encoding.Binary.to_bytes_exn Header.encoding block_header_data in
  let header : Block_header.t = { shell ; protocol_data = block_header_data_encoded } in
  let header_encoded = Data_encoding.Binary.to_bytes_exn Block_header.encoding header in
  let preapply_result = List.hd preapply_result in
  let operations = [ snd preapply_result.applied] in
  Shell_services.Injection.block cctxt header_encoded operations
  >>=? fun block_hash ->
  cctxt#message "Injected block %a" Block_hash.pp_short block_hash
  >>= fun () ->

First, it retrieves the applied operations from the mempool using Demo_block_services.Mempool.pending_operations. It then uses the pre-apply service and ask the node to build a block based on the proposed operations and block header data. The block is then encoded and sent to the node through the injection service.

Sample execution of the protocol

First we activate the protocol using protocol_parameters.json defined as

{'init_a': 100, 'init_b': 100}
# tezos-client -block genesis activate protocol ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT with fitness 1 and key activator and parameters protocol_parameters.json --timestamp 2019-07-05T14:30:35Z
Injected BLf2cXRZKsby

This bakes block of level 1, running protocol genesis, with demo_counter scheduled for the next block.

# tezos-client rpc get /chains/main/blocks/head/metadata
{ "protocol": "ProtoGenesisGenesisGenesisGenesisGenesisGenesk612im",
  "next_protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
  "test_chain_status": { "status": "not_running" }, "max_operations_ttl": 0,
  "max_operation_data_length": 100, "max_block_header_length": 100,
  "max_operation_list_length": [ { "max_size": 1000 } ] }

Although the head is a genesis block, demo_counter has already been activated and we can bake an empty block using the bake command from demo_counter client library.

# tezos-client bake '"This is block 2"'
Injected block BLrQqbn13Vrb

We can check that the block was baked properly, in particular the block header data has been set as expected. We can also see the protocol state State.t in the block metadata encoded as specified by State.encoding.

# tezos-client rpc get /chains/main/blocks/head/
{ "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
  "chain_id": "NetXdQprcVkpaWU",
  "hash": "BLrQqbn13VrbzUprxQypzAg6fc7YmsHaZvGwGrHJk8a4eG6e11B",
    { "level": 2, "proto": 1,
      "predecessor": "BLf2cXRZKsbygWJdtf1PBbrSg8yHkNK39bgoApvdbYBd1EX9ung",
      "timestamp": "2019-07-05T14:30:36Z", "validation_pass": 1,
      "fitness": [ "01", "0000000000000002" ],
      "context": "CoVpDgKDiWZ9xcodUFng1C8oGvfXEqBCD5XxQjB8Jrwptkx3vHUB",
      "demo_block_header_data": "This is block 2" },
    { "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
      "next_protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
      "test_chain_status": { "status": "not_running" },
      "max_operations_ttl": 0, "max_operation_data_length": 100,
      "max_block_header_length": 100,
      "max_operation_list_length": [ { "max_size": 1000 } ], "demo_a": 100,
      "demo_b": 100 }, "operations": [ [] ] }

We now inject three operations using client commands.

tezos-client incra
Operation receipt: operation applied successfully
Injected: op5gBsE7EMi7
# tezos-client incrb
Operation receipt: operation applied successfully
Injected: oo2YhBbAY8Vr
# tezos-client transfer 10
Operation receipt: operation applied successfully
Injected: opJFLuHR98tf

The operations are known to the node, they appear as applied in the node mempool.

# tezos-client rpc get /chains/main/mempool/pending_operations
{ "applied":
    [ { "hash": "op45sL79jASRf41kpL5NDDbAUnQeTfwgZpVnZi1sXy4Cj5x18m9",
        "branch": "BLa7SnHxjHqPTsGSE2fi8sHBm39u9g6Psd9qPZm4rJCqhzHdkSp",
        "IncrA": {} },
      { "hash": "opV6ZMR2z2ZZUSjetzPTknPisjN6x5eFCQnAWhuKNic6GsiRLW7",
        "branch": "BLa7SnHxjHqPTsGSE2fi8sHBm39u9g6Psd9qPZm4rJCqhzHdkSp",
        "IncrB": {} },
      { "hash": "oo1a3gwKnXFqaHuhpgMb5x69wm3mwbid4hn8Ry1iX8jbvXzQQs7",
        "branch": "BLa7SnHxjHqPTsGSE2fi8sHBm39u9g6Psd9qPZm4rJCqhzHdkSp",
        "Transfer": 10 } ], "refused": [], "branch_refused": [],
  "branch_delayed": [], "unprocessed": [] }

We bake the third block.

tezos-client bake '"This is block 3"'
Injected block BLz4SrcTnBQU

We can see now that the three operations appear in the operations section of the block, encoded as specified by Proto_operation.encoding. The receipt for each operation also appears in this section.

tezos-client rpc get /chains/main/blocks/head/
{ "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
  "chain_id": "NetXdQprcVkpaWU",
  "hash": "BLz4SrcTnBQUiXXks1FTzGR9d5MsX6F1mhZ2g23bHTcaQwJbk3S",
    { "level": 3, "proto": 1,
      "predecessor": "BLrQqbn13VrbzUprxQypzAg6fc7YmsHaZvGwGrHJk8a4eG6e11B",
      "timestamp": "2019-07-05T14:30:38Z", "validation_pass": 1,
      "fitness": [ "01", "0000000000000003" ],
      "context": "CoVXzytYqZcw4RQknAZJpK3RAFeLrzcZG2zDMdzuacpDPjX7YSor",
      "demo_block_header_data": "This is block 3" },
    { "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
      "next_protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
      "test_chain_status": { "status": "not_running" },
      "max_operations_ttl": 0, "max_operation_data_length": 100,
      "max_block_header_length": 100,
      "max_operation_list_length": [ { "max_size": 1000 } ], "demo_a": 91,
      "demo_b": 111 },
    [ [ { "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
          "chain_id": "NetXdQprcVkpaWU",
          "hash": "op5gBsE7EMi7gsR3xtSMMQms9XN8Pka5N1pT8XGuN1iP2siizkx",
          "branch": "BLrQqbn13VrbzUprxQypzAg6fc7YmsHaZvGwGrHJk8a4eG6e11B",
          "data": { "IncrA": {} },
            { "demo_operation_receipt": "operation applied successfully" } },
        { "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
          "chain_id": "NetXdQprcVkpaWU",
          "hash": "oo2YhBbAY8Vr2ASXj5k3PggXoxDVYWAQGvK1Sm22GhPFcXdvjQq",
          "branch": "BLrQqbn13VrbzUprxQypzAg6fc7YmsHaZvGwGrHJk8a4eG6e11B",
          "data": { "IncrB": {} },
            { "demo_operation_receipt": "operation applied successfully" } },
        { "protocol": "ProtoDemoCounterDemoCounterDemoCounterDemoCou4LSpdT",
          "chain_id": "NetXdQprcVkpaWU",
          "hash": "opJFLuHR98tfTTXGhsnMMjg6KxgVW8qx7LZ5q2nrhoySXySqZSS",
          "branch": "BLrQqbn13VrbzUprxQypzAg6fc7YmsHaZvGwGrHJk8a4eG6e11B",
          "data": { "Transfer": 10 },
            { "demo_operation_receipt": "operation applied successfully" } } ] ] }

We can finally test our two RPCs to query the counter values.

tezos-client rpc get /chains/main/blocks/head/counter/a
tezos-client rpc get /chains/main/blocks/head/counter/b

The node’s trace is similar to the one presented in the previous blog post. What we see in addition are three chunks of output of the form:

Jul  8 14:27:20 - demo-counter: begin_construction (mempool): pred_fitness = 01::0000000000000002  constructed fitness = 01::0000000000000003
Jul  8 14:27:20 - demo-counter: apply_operation
Jul  8 14:27:20 - demo-counter: finalize_block: fitness = 01::0000000000000003

Jul  8 14:27:20 - demo-counter: apply_operation
Jul  8 14:27:20 - demo-counter: apply_operation
Jul  8 14:27:20 - prevalidator.NetXdQprcVkpa.ProtoDemoCou_1: injecting operation op25ATABm3GS2AxZr8QFAaz9qor6JSqjYuYx4MSVXYtVZ86LaxF
Jul  8 14:27:20 - prevalidator.NetXdQprcVkpa.ProtoDemoCou_1: Pushed: 2019-07-08T12:27:20-00:00, Treated: 2019-07-08T12:27:20-00:00, Completed: 2019-07-08T12:27:20-00:00

The first 3 lines correspond to the call Demo_block_services.Helpers.Preapply.operations cctxt [op] and the next four lines to the call Shell_services.Injection.operation (both calls triggered by Client_proto_commands.inject_op). Finally, when the last block is created, we see the following output:

Jul  8 14:27:22 - demo-counter: begin_construction (block): pred_fitness = 01::0000000000000002  constructed fitness = 01::0000000000000003
Jul  8 14:27:22 - demo-counter: apply_operation
Jul  8 14:27:22 - demo-counter: apply_operation
Jul  8 14:27:22 - demo-counter: apply_operation
Jul  8 14:27:22 - demo-counter: finalize_block: fitness = 01::0000000000000003

Jul  8 14:27:22 - demo-counter: begin_application: pred_fitness = 01::0000000000000002  block_fitness = 01::0000000000000003
Jul  8 14:27:22 - demo-counter: apply_operation
Jul  8 14:27:22 - demo-counter: apply_operation
Jul  8 14:27:22 - demo-counter: apply_operation
Jul  8 14:27:22 - demo-counter: finalize_block: fitness = 01::0000000000000003
Jul  8 14:27:22 - validator.block: Block BMQFowYxF9WS6FjVRL2hMDEcV3LrViLFe9v4yo4W9tSPcyUq7Md successfully validated
Jul  8 14:27:22 - validator.block: Pushed: 2019-07-08T12:27:22-00:00, Treated: 2019-07-08T12:27:22-00:00, Completed: 2019-07-08T12:27:22-00:00

Here again, the chunks can be seen as being composed of two parts, one for Demo_block_services.Helpers.Preapply.block and one for Shell_services.Injection.block. In each of the parts apply_operation is called three times, once for each operation included in the block.

This scenario can be reproduced using the following python test (this requires to install Python tools and libraries, as described here. It launches a node, and runs the client commands to activate the protocol and interact with it.

cd tests_python
mkdir tmp
pytest -s tests/ --log-dir=tmp

The node trace appears in tmp/node0_0.txt.


We presented a simple protocol demo_counter which explores further the interface between the shell and the protocol, and uses more features available to the protocol developer such as RPC services. Besides, this protocol comes with a library that extends tezos-client with new commands to interact with the protocol.

In the next post, we will present demo_account, an account-based protocol, where transactions and blocks must be signed in order to be valid.

Formally Verifying a Critical Smart Contract

We present the formal verification of the Spending Limit Contract, a critical component of the Cortez wallet.

One of the main goals of Nomadic Labs is the development and applications of formal methods in the domain of distributed software, blockchains and smart contracts. In particular for the Tezos blockchain, for which we also develop the Cortez smartphone wallet (Android, iPhone). This wallet helps Tezos users manage their account and funds in a safe and secure manner. How can the user be...

Read More
Sapling integration in Tezos - Tech Preview

We are happy to announce a first technology preview of our integration in Tezos of the core of the Sapling protocol developed by the ZCash project. By extending the Michelson smart contract language, this work allows for the exchange of digital assets in a privacy preserving way. Why Sapling? In recent years, we’ve seen much progress towards enabling privacy-preserving payments on public ledgers, both in academic research and in the real world deployement with projects such as Zcash, Monero, or Aztec. In...

Read More
A new reward formula for Carthage

Note: This analysis was done with the help of Arthur Breitman and Bruno Blanchet (Inria). The code used for the analysis can be found at this url. A new reward formula for Carthage In this article, we present a new reward formula that we propose for inclusion in Carthage. This new formula is designed to make the network more robust to non-cooperative baking strategies. It does so without changing the total amount of rewards earned by bakers...

Read More
Michelson updates in 005

Changes in Michelson As hinted at in a previous blog post, we’ve been working on improving different parts of the protocol, including our favourite smart contract language: Michelson. The changes made to Michelson in this proposal intend to simplify smart contract development by making the code of complex contracts simpler and cleaner. In particular: smart contracts now support entrypoints contracts can now create, store and transmit as many big_maps as they want comparable types are now closed under products (i.e. the pair constructor) a new instruction, CHAIN_ID, allows...

Read More
Analysis of Emmy+

19/11/2019 update: Following feedback from Daniel Moroz and Michael Neuder, we edited the blog post in order to nuance our previous statement that “selfish baking is not profitable”. More precisely, selfish baking only gives tiny profits, for instance 0.0016% gains with respect to the honest rewards, for a dishonest baker with 30% stake fraction. A new table gives the details. Note: This analysis was done with the help of Arthur Breitman and Bruno Blanchet (Inria). The...

Read More
How to write a Tezos protocol

A Tezos node is parameterized by a software component called an economic protocol (or protocol for short). Different protocol implementations can be used to implement different types of blockchains. This is the first post of a tutorial series on how to implement such a protocol. We will see how to write, compile, register, activate and use an extremely simple protocol. By doing so, we will also start to explore the interface between the protocol and the node (more specifically the shell component of...

Read More
Athens: Our Proposals for the First Voted Amendment

This blog post is a preview of Athens: our protocol proposal for the first voted upgrade of Tezos. As announced in the last meanwhile at Nomadic, we shall propose two upgrades: one lowers the roll size to 8,000 tez, the other leaves it unchanged at 10,000 tez. Both alternatives will include an increase of the gas limit. The hashes of both versions will be proposed on mainnet later this week, now that a new proposal period has begun. Later this week, we will publish a...

Read More
Amendments at Work in Tezos

We are now on the verge of submitting a protocol upgrade to a vote, and it seems like a good opportunity to explain in details the way in which Tezos node handles amendment in practice. Brace yourselves, this article is quite technical, as are all articles in our in-depth category. Still, as we did in the previous one on snapshots, we’ll try to explain the stakes and announcements and give a brief summary in a short foreword understandable even by non-programmers. The original whitepaper...

Read More
Introducing Snapshots and History Modes for the Tezos Node

In this article, we introduce two new features for the Tezos node: snapshots and history modes. A snapshot is a file that contains everything necessary to restore the state of a node at a given block. A node restored via a snapshot can synchronise and help other nodes synchronise in the existing network. The only difference is that you cannot query the chain context (balances, baking rights, etc.) before the restoration point, but you can still get the full chain history. In conjunction, we also...

Read More

Receive Updates