[Utility] Introduce on-chain Actor (un)Jailing

[Olshansk]

Objective

Design & implement a framework of how to handle on-chain (un)jailing with placeholders for the triggers & events to be determined at a future point in time.

Origin Documents

• Cosmos SDK documentation on jailing/unjailing
• Morse (pocket-core) has some references to unjailing which can be found here

Goals

• Jailing - a state that an on-chain actor can be in which prevents it from being able to participate in any on-chain operations
• Unjailing - the process of re-enable a previously jailed actor in resuming normal on-chain operations
• Slashing actors that have been jailed for a sufficiently long period of time
• Evaluate options for how on-chain actor (un)jailing could be implemented
• Design, document and implement (un)jailing of Applications, Suppliers & Gateways
• Create a pattern for reusable entrypoints that’ll trigger actor jailing
  • If new functionality is added for an actor, it should be “easy” to make sure it is not available while the actor is jailed

Deliverables

IMPORTANT NOTE TO THE OWNER: The deliverables below are not a complete list of everything that needs to be done, but rather a starting point to help. The research, which is part of the deliverables, should drive most of the design decisions.

• Research
  • [ ] Put together a reference doc with bullet points, links, highlights, learnings capturing the research below(< 1 page total)
  • Other Projects
    • [ ] Investigate how validator (un)jailing is designed and implemented in the Cosmos SDK
    • [ ] Investigate how at least one other project does (non-validator) actor jailing in the Cosmos ecosystem
  • Morse
    • [ ] Investigate how actor (un)jailing currently works in Morse
• Before implementation
  • [ ] Prepare a 1-pager “mini spec” notion / docs / markdown and get feedback & review before kicking off the implementation
• Implementation requirements
  • [ ] Time-based unjailing should follow the same pattern as in unbonding (#489)
  • [ ] Authority (governance address) should be able to jail / unjail any actor
  • [ ] [TBD] Unjailing should happen automatically N blocks since the moment the actor was jailed
    • [ ] N should have a unique governance parameter for every actor: Application, Supplier, Gateway
  • [ ] [TBD] If an actor is jailed for M blocks, it will get slashed by some amount Z
  • [ ] Create custom N , M , Z governance params unique to each actor
• During implementation
  • [ ] Separate PRs per actor after the foundation business logic is in place
  • [ ] Add unit / integration tests
  • [ ] Add at least 1 E2E test
  • [ ] Make target to trigger the E2E test
  • [ ] Make targets to execute governance / auth trigger jailing / unjailing of a specific address
• After implementation
  • [ ] Add a new page in docusurus that explains how the implemented mechanism works
  • [ ] Create a follow-up ticket to add the jailing / unjailing triggering business logic

Non-goals / Non-deliverables

• Design or determine the automatic (on-chain, not manually triggered) events that cause jailing

General deliverables

• [ ] Comments: Add/update TODOs and comments alongside the source code so it is easier to follow.

---

Creator: @olshansk Co-Owners: @moatus

[Olshansk]

@moatus Could you review the requirements for unstaking here. I'm thinking of mounting out the engineering work for it.