Node doesn't start with `archive-canonical` flag

[vedhavyas]

Is there an existing issue?

• I have searched the existing issues

Experiencing problems? Have you tried our Stack Exchange first?

• This is not a support question.

Description of bug

The node errors out when I start with archive-canonical. Our substrate fork is not too old and be found here - https://github.com/subspace/substrate.

subspace-node-1    | 2022-12-01 16:52:02.890  INFO main sc_cli::runner: Subspace    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: ✌️  version 0.1.0-unknown    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: ❤️  by Subspace Labs <https://subspace.network>, 2021-2022    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: 📋 Chain specification: Subspace Gemini 3a    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: 🏷  Node name: ved_local    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: 👤 Role: AUTHORITY    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: 💾 Database: ParityDb at /var/subspace/chains/subspace_gemini_3a/paritydb/full    
subspace-node-1    | 2022-12-01 16:52:02.891  INFO main sc_cli::runner: ⛓  Native runtime: subspace-0 (subspace-0.tx0.au0)    
subspace-node-1    | 2022-12-01 16:52:03.288  INFO main sc_service::client::client: [PrimaryChain] 🔨 Initializing Genesis block/state (state: 0x1c3c…f9bb, header-hash: 0x8797…5b2f)    
subspace-node-1    | Error: SubstrateService(Other("Failed to build a full subspace node: Sub(Client(RuntimeApiError(Application(UnknownBlock(\"State already discarded for 0x8797298eb8cd5c3b3a08b6c1a6ed7809ee40e7f80fc0acf3cdb3fde00a435b2f\")))))"))

If I run the node with archive or other pruning method. it works. The Node is completely pruned everytime.
Also note that hash that is missing is the Genesis block.

Steps to reproduce

This is the docker-compose I used.

services:
  node:
    image: ghcr.io/subspace/node:gemini-3a-2022-dec-01-aarch64
    volumes:
      - node:/var/subspace:rw
    ports:
      - "0.0.0.0:30333:30333"
    restart: unless-stopped
    command: [
      "--chain", "gemini-3a",
      "--base-path", "/var/subspace",
      "--execution", "wasm",
      "--state-pruning", "archive-canonical",
      "--blocks-pruning", "archive-canonical",
      "--port", "30333",
      "--rpc-cors", "all",
      "--rpc-methods", "safe",
      "--unsafe-ws-external",
      "--validator",
      "--log=subspace=debug",
      "--name", "node_local"
    ]
    healthcheck:
      timeout: 5s
      interval: 30s
      retries: 5

Executables are also fond here - https://github.com/subspace/subspace/releases/tag/gemini-3a-2022-dec-01

[maciejnems]

Recently I tried running a substrate based node with --state-pruning archive-canonical and got the same error.

[Polkadot-Forum]

This issue has been mentioned on Polkadot Forum. There might be relevant details there:

https://forum.polkadot.network/t/polkadot-release-analysis-v0-9-39/2277/1