Skip to content
This repository has been archived by the owner on Mar 5, 2020. It is now read-only.

Latest commit

 

History

History
64 lines (47 loc) · 2.63 KB

INSTRUCTIONS.md

File metadata and controls

64 lines (47 loc) · 2.63 KB
Raw

These are detailed instructions for launching a mainnet

If you are appointed Interim Block Producer, here is our experience and tips:

  1. In the sample_config, tweak hook_boot_node.sh to reflect your environment.

  2. Run eos-bios boot --single.

  3. This will generate a genesis.json and genesis.pub and genesis.key.

  4. If you ever need to restart the chain because injection fails or whatever, run: eos-bios boot --single --reuse-genesis. This way, you don't need to disseminate another genesis.json and require everyone to clear-up their storage.

  5. The node your boot NEEDS --max-transaction-time=5000 for transactions not to fail.

  6. You should not mesh to anyone during injection. There are already bots that are sending automated transactions.. if you are meshed with someone exposing an API endpoint, the boot sequence will surely be impacted, and validation will most probably fail.

  7. Once injection succeeds, and validation passes (the eos-bios built-in validation) you can enable author-whitelist = nobody in the configuration, and add a few p2p-peer-address statements. Do not use contract-whitelist unless #3418 has been merged and you are running updated code.

  8. Restart your node, and resume production if it was paused on startup (remove the --genesis-json flag).

  9. Blocks should start propagating to the network, but no transaction can make it into the chain.

  10. Once everyone got the contents, they should be able to validate it without issue, as it is read-only.

The community might want to publish reports of the different validation tools so that everyone gains confidence it is the chain we want to bring to the world.

  1. When the time comes to unlock the chain for transaction, stop the node, remove author-whitelist

Emergency handoff

If your node is unable to produce, its connectivty breaks, or you want to move that producing node off of your Raspberry Pi, do the following:

  • Call /v1/producer/pause on the first node.
  • Ensure the node to which you are transitioning was fully sync'd and has the last blocks
  • Make sure enable-stale-production = true on the target node.
  • Call /v1/producer/resume on the new node or restart without --pause-on-startup.
  • WARN: do NOT hand it off back to the INITIAL node without a PROPER RESTART OF THE FIRST NODE. Otherwise, you will hit: EOSIO/eos#3442
  • If you handoff before unfreezing the chain, make sure the target node has author-whitelist = nobody also, otherwise you risk unfreezing the chain too early.