-
Notifications
You must be signed in to change notification settings - Fork 766
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Staking] Currency <> Fungible migration #5501
base: master
Are you sure you want to change the base?
Conversation
balance hold checks both frozen and reserved wip: around 25 tests failing check Holds instead of locks 20 tests failing fmt 11 fails 4 fails 2 failing 1 fail all tests pass but pending a hygiene check of code fix compile minor refactor remove T::Currency calls from asset mod
…staking-migrate-currency-to-fungible-2
// Get rid of the extra consumer we used to have with OldCurrency. | ||
frame_system::Pallet::<T>::dec_consumers(&stash); | ||
|
||
Self::deposit_event(Event::<T>::CurrencyMigrated { stash: stash.clone(), force_withdraw }); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this event suppossed to notify of how much funds were migrated?
Because by looking at the if-else above it looks like it's more like funds_still_locked_in_old_currency.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not sure if I understand funds_still_locked_in_old_currency
? Once we get here, there should be nothing locked with old currency.
The event is meant to notify clients that the given account has migrated to new currency along with the information if their ledger has reduced by some balance because of this operation.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I understood that the "else" condition is when someone has more funds locked for staking than in their free balance thus we cannot "hold" as much as is needed so we go just for as much as possible.
Regarding funds_still_locked_in_old_currency
, I called it so because in if
zero value is returned and in else
it's the difference between staked and free balance and since the free balance will be held then the result is what's left to unlock and move to hold.
Is my reasoning correct?
@@ -89,13 +93,27 @@ pub mod pallet { | |||
|
|||
#[pallet::config(with_default)] | |||
pub trait Config: frame_system::Config { | |||
/// The old trait for staking balance. Deprecated and only used for migrating old ledgers. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shouldn't the pallet be annotated with deprecated then?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Not pallet but yeah may be config item can be annotated as deprecated (if its possible).
/// | ||
/// Does not include the current stake. | ||
pub fn free_to_stake<T: Config>(who: &T::AccountId) -> BalanceOf<T> { | ||
// since we want to be able to use frozen funds for staking, we force the reduction. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The note was related to the usage of Fortitude::Force
. Any funds that is locked or frozen (such as for governance or vesting) should still be usable for staking.
let _ = T::Currency::set_balance(who, value - staked_balance + ed); | ||
} else { | ||
// else reduce the staked balance. | ||
update_stake::<T>(who, value).expect("can remove from what is staked"); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
update_stake::<T>(who, value).expect("can remove from what is staked"); | |
update_stake::<T>(who, value).expect("cannot remove from what is staked"); |
#[test] | ||
fn ledger_update_creates_hold() { | ||
ExtBuilder::default().has_stakers(true).build_and_execute(|| { | ||
// GIVEN alice who is a nominator with old currency |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
// GIVEN alice who is a nominator with old currency | |
// GIVEN alice who is a nominator with new currency |
// We should never have more than expected providers. | ||
ensure!(actual_providers <= expected_providers, Error::<T>::BadState); | ||
|
||
// if actual provider is less than expected, it is already migrated. | ||
ensure!(actual_providers == expected_providers, Error::<T>::AlreadyMigrated); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this means it should be just equal. first ensure is redundant
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The errors will be different. If greater -> bad state, if less -> AlreadyMigrated.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
but you never get AlreadyMigrated
error
let expected_providers = | ||
// provider is expected to be 1 but someone can always transfer some free funds to | ||
// these accounts, increasing the provider. | ||
if asset::free_to_stake::<T>(&stash) >= asset::existential_deposit::<T>() { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
should not be balance >= ed
? free_to_stake
does not include the ED
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These (keyless) accounts had a explicit provider added to them by delegated staking.
When you have the extra provider, free_to_stake
will include ED. Retrying to migrate will always fail as actual provider should be 0, always less than expected providers.
The test hold_migration::virtual_staker_consumer_provider_dec
checks these scenarios.
@@ -262,7 +264,8 @@ impl<T: Config, Staking: StakingInterface<Balance = BalanceOf<T>, AccountId = T: | |||
pool_account: Pool<Self::AccountId>, | |||
_: Member<Self::AccountId>, | |||
) -> BalanceOf<T> { | |||
T::Currency::balance(&pool_account.0).saturating_sub(Self::active_stake(pool_account)) | |||
// free/liquid balance of the pool account. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
balance()
contains the frozen part which means can be not liquid
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Technically this works because pool will never have frozen funds (its a pool generated pot account) but is confusing. I will fix it.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
seems fixed now?
@Ank4n can you explain why force withdrawal not just 59 DOT? so locks are not there? |
Yes we had a bug that caused ledgers to be overwritten. It didn't exactly allow anyone to stake more than their balance, but it overwritten their controller account with stash account's stake. Most of them have been fixed here: polkadot-fellows/runtimes#447. |
All GitHub workflows were cancelled due to failure one of the required jobs. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me. But consider this approve as from someone who is not an expert in the staking topic
@@ -2020,6 +2020,7 @@ pub mod pallet { | |||
ensure!(!PoolMembers::<T>::contains_key(&who), Error::<T>::AccountBelongsToOtherPool); | |||
|
|||
let mut bonded_pool = BondedPool::<T>::get(pool_id).ok_or(Error::<T>::PoolNotFound)?; | |||
ensure!(bonded_pool.state == PoolState::Open, Error::<T>::NotOpen); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
eh, seems like a bug from the past? let's add a unit test.
impl DelegationInterface for DelegateMock { | ||
type Balance = Balance; | ||
type AccountId = AccountId; | ||
fn agent_balance(agent: Agent<Self::AccountId>) -> Option<Self::Balance> { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this is basically the "staked, and active in staking (minus pending slash)" balance?
} | ||
|
||
fn remove_agent(agent: Agent<Self::AccountId>) -> DispatchResult { | ||
let mut agents = AgentBalanceMap::get(); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
probably not needed for tests but to simulate the real code better, we should error if agent
is in fact not an agent?
Or assert!(agents.contains(agent.get()))
|
||
let agent = agent.get(); | ||
let mut agents = AgentBalanceMap::get(); | ||
agents.get_mut(&agent).map(|(d, _, _)| *d += amount); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
agents.get_mut(&agent).map(|(d, _, _)| *d += amount); | |
agents.get_mut(&agent).map(|(d, _, _)| *d += amount).ok_or(/* some dispatch error */); |
@@ -3925,6 +3861,10 @@ mod withdraw_unbonded { | |||
let mut member = PoolMember { pool_id: 1, points: 10, ..Default::default() }; | |||
PoolMembers::<Runtime>::insert(11, member.clone()); | |||
|
|||
// set agent and delegator balance |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
How is it that these two were not needed before this change?
@@ -1331,11 +1344,12 @@ fn disable_pool_operations_on_non_migrated() { | |||
assert_ok!(Pools::migrate_pool_to_delegate_stake(RuntimeOrigin::signed(10), 1)); | |||
assert_eq!( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This file had really good tests, I learned a lot by reading them!
PR is in good shape, and thanks to incremental code, the changes are easy to digest, good work!
|
Migrate staking currency from
traits::LockableCurrency
totraits::fungible::holds
.Resolves part of #226.
Changes
Nomination Pool
TransferStake is now incompatible with fungible migration as old pools were not meant to have additional ED. Since they are anyways deprecated, removed its usage from all test runtimes.
Staking
Currency
becomes of typeFungible
whileOldCurrency
is theLockableCurrency
used before.migrate_currency()
releases the oldlock
along with some housekeeping.Delegated Staking
The pallet does not add provider for agents anymore.
Migration stats
Polkadot
Total accounts that can be migrated: 59564
Accounts failing to migrate: 0
Accounts with stake force withdrawn greater than ED: 59
Total force withdrawal: 29591.26 DOT
Kusama
Total accounts that can be migrated: 26311
Accounts failing to migrate: 0
Accounts with stake force withdrawn greater than ED: 48
Total force withdrawal: 1036.05 KSM
Full logs here.
Note about locks (freeze) vs holds
With locks or freezes, staking could use total balance of an account. But with holds, the account needs to be left with at least Existential Deposit in free balance. This would also affect nomination pools which till now has been able to stake all funds contributed to it. An alternate version of this PR is #5658 where staking pallet does not add any provider, but means pools and delegated-staking pallet has to provide for these accounts and makes the end to end logic (of provider and consumer ref) lot less intuitive and prone to bug.
This PR now introduces requirement for stakers to maintain ED in their free balance. This helps with removing the bug prone incrementing and decrementing of consumers and providers.
TODO
Call::restore_ledger
be removed? @gpestanamigrate_currency
.Followup