Enable async backing for system parachains: Phase 1 #228
Conversation
|
@georgepisaltu could you take a look too? |
|
Are there any node equivalent changes ? And are there any dependencies between them, do they need to be deployed in a certain order ? |
|
@alexggh Yes, for both questions. This is phase 1; that is, it doesn't change any behavior of the runtime or node, but it exposes new interfaces (namely Phase 2 will involve updating the node to use those interfaces (that is done in paritytech/polkadot-sdk#3630; testing and description are in progress). After phase 1 is enacted, we will merge phase 2 and start upgrading the collators. After that is done, everything will be async backing ready but still running at 12-sec blocktimes. After that, at phase 3, one more update to the parachain runtimes will be made, bumping constants (slot duration, capacity, and velocity) to enable async backing with 6-sec blocktimes. |
Co-authored-by: Bastian Köcher <[email protected]>
|
/merge |
|
Enabled Available commands
For more information see the documentation |
Failed to update PR ❌There was an error while trying to keep this PR You may have conflicts More info in the logs 📋 |
Failed to update PR ❌There was an error while trying to keep this PR You may have conflicts More info in the logs 📋 |
…m-parachains-phase-1
Head branch was pushed to by a user without write access
|
/merge |
There was a problem running the action.❌😵❌ Please find more information in the logs. |
|
Seems like it requires a re-review after manually merging |
|
/merge |
|
Enabled Available commands
For more information see the documentation |
fellowship-merge-bot
bot
merged commit c5f8556
into
polkadot-fellows:main
#3630) This is phase 2 of async backing enablement for the system parachains on the production networks. ~~It should be merged after polkadot-fellows/runtimes#228 is enacted. After it is released,~~ all the system parachain collators should be upgraded, and then we can proceed with phase 3, which will enable async backing in the runtimes. UPDATE: Indeed, we don't need to wait for the runtime upgrade enactions. The lookahead collator handles the transition by itself, so we can upgrade ASAP. ## Scope of changes Here, we eliminate the dichotomy of having "generic Aura collators" for the production system parachains and "lookahead Aura collators" for the testnet system parachains. Now, all the collators are started as lookahead ones, preserving the logic of transferring from the shell node to Aura-enabled collators for the asset hubs. So, indeed, it simplifies the parachain service logic, which cannot but rejoice.
This PR aims to prepare system parachains' runtimes for the async backing enabling. Namely, it:
AuraUnincludedSegmentApifor every runtime;