- Apr 26, 2024
-
-
Tsvetomir Dimitrov authored
Closes https://github.com/paritytech/polkadot-sdk/issues/1966, https://github.com/paritytech/polkadot-sdk/issues/1963 and https://github.com/paritytech/polkadot-sdk/issues/1962. Disabling strategy specification [here](https://github.com/paritytech/polkadot-sdk/pull/2955). (Updated 13/02/2024) Implements: * validator disabling for a whole era instead of just a session * no more than 1/3 of the validators in the active set are disabled Removes: * `DisableStrategy` enum - now each validator committing an offence is disabled. * New era is not forced if too many validators are disabled. Before this PR not all offenders were disabled. A decision was made based on [`enum DisableStrategy`](https://github.com/paritytech/polkadot-sdk/blob/bbb6631641f9adba30c0ee6f4d11023a424dd362/substrate/primitives/staking/src/offence.rs#L54). Some offenders were disabled for a whole era, some just for a session, some were not disabled at all. This PR changes the disabling behaviour. Now a validator committing an offense is disabled immediately till the end of the current era. Some implementation notes: * `OffendingValidators` in pallet session keeps all offenders (this is not changed). However its type is changed from `Vec<(u32, bool)>` to `Vec<u32>`. The reason is simple - each offender is getting disabled so the bool doesn't make sense anymore. * When a validator is disabled it is first added to `OffendingValidators` and then to `DisabledValidators`. This is done in [`add_offending_validator`](https://github.com/paritytech/polkadot-sdk/blob/bbb6631641f9adba30c0ee6f4d11023a424dd362/substrate/frame/staking/src/slashing.rs#L325) from staking pallet. * In [`rotate_session`](https://github.com/paritytech/polkadot-sdk/blob/bdbe98297032e21a553bf191c530690b1d591405/substrate/frame/session/src/lib.rs#L623) the `end_session` also calls [`end_era`](https://github.com/paritytech/polkadot-sdk/blob/bbb6631641f9adba30c0ee6f4d11023a424dd362/substrate/frame/staking/src/pallet/impls.rs#L490) when an era ends. In this case `OffendingValidators` are cleared **(1)**. * Then in [`rotate_session`](https://github.com/paritytech/polkadot-sdk/blob/bdbe98297032e21a553bf191c530690b1d591405/substrate/frame/session/src/lib.rs#L623) `DisabledValidators` are cleared **(2)** * And finally (still in `rotate_session`) a call to [`start_session`](https://github.com/paritytech/polkadot-sdk/blob/bbb6631641f9adba30c0ee6f4d11023a424dd362/substrate/frame/staking/src/pallet/impls.rs#L430 ) repopulates the disabled validators **(3)**. * The reason for this complication is that session pallet knows nothing abut eras. To overcome this on each new session the disabled list is repopulated (points 2 and 3). Staking pallet knows when a new era starts so with point 1 it ensures that the offenders list is cleared. --------- Co-authored-by:
ordian <[email protected]> Co-authored-by:
ordian <[email protected]> Co-authored-by:
Maciej <[email protected]> Co-authored-by:
Gonçalo Pestana <[email protected]> Co-authored-by:
Kian Paimani <[email protected]> Co-authored-by: command-bot <> Co-authored-by:
Ankan <[email protected]>
-
- Nov 01, 2023
-
-
Ankan authored
helps https://github.com/paritytech/polkadot-sdk/issues/439. closes https://github.com/paritytech/polkadot-sdk/issues/473. PR link in the older substrate repository: https://github.com/paritytech/substrate/pull/13498. # Context Rewards payout is processed today in a single block and limited to `MaxNominatorRewardedPerValidator`. This number is currently 512 on both Kusama and Polkadot. This PR tries to scale the nominators payout to an unlimited count in a multi-block fashion. Exposures are stored in pages, with each page capped to a certain number (`MaxExposurePageSize`). Starting out, this number would be the same as `MaxNominatorRewardedPerValidator`, but eventually, this number can be lowered through new runtime upgrades to limit the rewardeable nominators per dispatched call instruction. The changes in the PR are backward compatible. ## How payouts would work like after this change Staking exposes two calls, 1) the existing `payout_stakers` and 2) `payout_stakers_by_page`. ### payout_stakers This remains backward compatible with no signature change. If for a given era a validator has multiple pages, they can call `payout_stakers` multiple times. The pages are executed in an ascending sequence and the runtime takes care of preventing double claims. ### payout_stakers_by_page Very similar to `payout_stakers` but also accepts an extra param `page_index`. An account can choose to payout rewards only for an explicitly passed `page_index`. **Lets look at an example scenario** Given an active validator on Kusama had 1100 nominators, `MaxExposurePageSize` set to 512 for Era e. In order to pay out rewards to all nominators, the caller would need to call `payout_stakers` 3 times. - `payout_stakers(origin, stash, e)` => will pay the first 512 nominators. - `payout_stakers(origin, stash, e)` => will pay the second set of 512 nominators. - `payout_stakers(origin, stash, e)` => will pay the last set of 76 nominators. ... - `payout_stakers(origin, stash, e)` => calling it the 4th time would return an error `InvalidPage`. The above calls can also be replaced by `payout_stakers_by_page` and passing a `page_index` explicitly. ## Commission note Validator commission is paid out in chunks across all the pages where each commission chunk is proportional to the total stake of the current page. This implies higher the total stake of a page, higher will be the commission. If all the pages of a validator's single era are paid out, the sum of commission paid to the validator across all pages should be equal to what the commission would have been if we had a non-paged exposure. ### Migration Note Strictly speaking, we did not need to bump our storage version since there is no migration of storage in this PR. But it is still useful to mark a storage upgrade for the following reasons: - New storage items are introduced in this PR while some older storage items are deprecated. - For the next `HistoryDepth` eras, the exposure would be incrementally migrated to its corresponding paged storage item. - Runtimes using staking pallet would strictly need to wait at least `HistoryDepth` eras with current upgraded version (14) for the migration to complete. At some era `E` such that `E > era_at_which_V14_gets_into_effect + HistoryDepth`, we will upgrade to version X which will remove the deprecated storage items. In other words, it is a strict requirement that E<sub>x</sub> - E<sub>14</sub> > `HistoryDepth`, where E<sub>x</sub> = Era at which deprecated storages are removed from runtime, E<sub>14</sub> = Era at which runtime is upgraded to version 14. - For Polkadot and Kusama, there is a [tracker ticket](https://github.com/paritytech/polkadot-sdk/issues/433) to clean up the deprecated storage items. ### Storage Changes #### Added - ErasStakersOverview - ClaimedRewards - ErasStakersPaged #### Deprecated The following can be cleaned up after 84 eras which is tracked [here](https://github.com/paritytech/polkadot-sdk/issues/433). - ErasStakers. - ErasStakersClipped. - StakingLedger.claimed_rewards, renamed to StakingLedger.legacy_claimed_rewards. ### Config Changes - Renamed MaxNominatorRewardedPerValidator to MaxExposurePageSize. ### TODO - [x] Tracker ticket for cleaning up the old code after 84 eras. - [x] Add companion. - [x] Redo benchmarks before merge. - [x] Add Changelog for pallet_staking. - [x] Pallet should be configurable to enable/disable paged rewards. - [x] Commission payouts are distributed across pages. - [x] Review documentation thoroughly. - [x] Rename `MaxNominatorRewardedPerValidator` -> `MaxExposurePageSize`. - [x] NMap for `ErasStakersPaged`. - [x] Deprecate ErasStakers. - [x] Integrity tests. ### Followup issues [Runtime api for deprecated ErasStakers storage item](https://github.com/paritytech/polkadot-sdk/issues/426 ) --------- Co-authored-by:
Javier Viola <[email protected]> Co-authored-by:
Ross Bulat <[email protected]> Co-authored-by: command-bot <>
-