SDK 0.25 / Gaia-9000 release checklist

[cwgoes]

Blocks:

• Simulation fixes for distribution module (@cwgoes / @rigelrozanski) R4R: Simulate transactions for fee distribution, simulate inflation, and fix a multitude of bugs discovered in the process of doing so #2501

Good to get in:

• Vesting accounts (@alexanderbez) [R4R]: Vesting Account(s) Implementation #2694 (currently blocked on final implementation detail)

Necessary for GoS but not gaia-9000:

• Import/export fix (@cwgoes) R4R: Fix state export/import, add to CI #2690

Completed:

• Inflation distribution & commission (@rigelrozanski) R4R: Minting #2527
• GenTx -> StdTx refactor (@alessio) WIP: Replace GenesisTx with StdTx #2422
• Removed validators are still slashable (@cwgoes / @sunnya97) Ensure that unbonding-delegations/redelegations for removed validators are still slashable #1673
• Faulty slashing when ex-validator is reintroduced into the validator set (@cwgoes) Faulty slashing when ex-validator is reintroduced into the validator set #1867
• Piggy bank implementation (@rigelrozanski) R4R: Distr-PR-5 Implement Distribution #2236
• Validator slashing logic (@cwgoes ) Failed to slash a validator for double sign at the first block #2372
• Validator power rankings (@ValarDragon) x/staking ValidatorPowerRank will have incorrect orderings #2439
• Add redelegation index (@cwgoes / @ValarDragon) x/stake: Add global redelegation/unbonding index #1402
• Validator Queues (@sunnya97) Implement Validator Queue #2412
• Redegation/Unbonding (@sunnya97) Implement Unbonding/Redelegation Queue #2393
• Validator unbonding state (R4R: Validator unbonding state #2163)
• Slashing period spec & implementation (R4R: Update slashing spec for slashing-by-period #2001 , R4R: Implement slashing period #2122)
• Piggy bank fee distribution spec (R4R: Piggy-bank distribution spec #1944)
• Transaction simulation, split generate/sign/broadcast (Generate and broadcast messages with multiple signatures via CLI #2204)
• Fix staking bugs (x/staking: A sequence of transactions causes one to delete a non-existent validator #2189, Invalid validator state in simulation #2277)
• Simple flat fee set in config.toml (keys: Should support ed25519 #1861)
• Update to Tendermint 0.24 (R4R: Update to Tendermint 0.24 (except NextValSet offsets) #2219)
• Debug simulation invariant failure (R4R: Add loose/bonded token supply invariants back to simulation #2162)
• Slashing changes for NextValSet (R4R: Offset validator set power calculations by one block #2255)
• Debug faults caught in simulation (Randomized fuzzer: slashing period ended before infraction #2224, Randomized fuzzer: expected to find slashing period, but none was found  #2225)

[ValarDragon]

I thought tx simulation was done for now. (Were unsure if it's necessary as we haven't made a decision for how accurately were doing gas at launch)

[alessio]

One question from me: should the simulation be turned on by default?

…

On Mon, Sep 3, 2018 at 7:10 PM, Dev Ojha ***@***.***> wrote: I thought tx simulation was done for now. (Were unsure if it's necessary as we haven't made a decision for how accurately were doing gas at launch) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#2220 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAN_7AxIruu8iAPK4pL27GG-EmzFrMIdks5uXWJpgaJpZM4WXxi8> .

-- Alessio Treglia | [email protected] 0416 0004 A827 6E40 BB98 90FB E8A4 8AE5 311D 765A

[ValarDragon]

I'd opt to leave it off for default until we make it not depend on privkey. If we enable it, we'll have to explain the UI to everyone using ledgers. (Plus my fears of side channel leakage only get exacerbated, especially since our underlying secp library isn't constant time)

[ValarDragon]

Also weren't we aiming for SDK v0.25 by end of this week? If so, this feels a bit optimistic to me. (Like the ABCI codespace thing probably won't happen by end of this week, I'm unsure of the status of most of the purely SDK features)

[cwgoes]

Also weren't we aiming for SDK v0.25 by end of this week? If so, this feels a bit optimistic to me. (Like the ABCI codespace thing probably won't happen by end of this week, I'm unsure of the status of most of the purely SDK features)

Possibly, maybe we should target Monday/Tuesday. When do you expect the ABCI codespace & event updates to be merged upstream?

[jackzampolin]

I think we are targeting monday/tuesday fwiw

[cwgoes]

I think we are targeting monday/tuesday fwiw

Also, I think we'll cut the ABCI updates from this release target since they aren't in TM 0.24.

[jackzampolin]

Looking more like an end of week (thursday/friday) release after Standup today.

[cwgoes]

All of #2277, #2241, #2264, #2225, and #2224 block this release.

[ValarDragon]

Can we update the iavl version for next release? #2281

[cwgoes]

Can we update the iavl version for next release? #2281

Let's do. Reviewed & merged.

[cwgoes]

Also we need to include #2210 in this release.

[alexanderbez]

and #2326?

[cwgoes]

Blocked also on #2396.

[cwgoes]

Also blocked on #2372.

[cwgoes]

Also blocked on #2162.

[cwgoes]

Also blocked on #2422.

[cwgoes]

Remaining simulation/staking/slashing concerns not addressed by #2394:

• Randomized fuzzer: expected to find slashing period, but none was found  #2225
• Randomized fuzzer: slashing period ended before infraction #2224
• Failed to slash a validator for double sign at the first block #2372
• simulation: Change parameters to get more slashes #2396

[alexanderbez]

Awesome -- thanks for compiling that @cwgoes. It'll probably be best if we readdress those once the refactor (#2394) has been merged.

[jackzampolin]

Updated the issue to reflect what needs doing.

[cwgoes]

I think I can also get to #1867 before GoS, since the other staking/slashing issues are done now.

Added to the list above.

[cwgoes]

I think #1673 has to block GoS, it's too much of an attack vector.

[jackzampolin]

Closing this issue as we have cut the 0.25 release