-
Notifications
You must be signed in to change notification settings - Fork 1.6k
Remove service, migrate all to service-new #1630
Conversation
- Move the chain-spec files to node-service - update sufficient cargo files that polkadot-service-new builds - not everything else builds yet
There's a lot more to go, but some progress has happened.
This begins to make Polkadot usable from Cumulus.
This reverts commit 7b8f9ba.
This is necessary because one of the current errors when building the test service boils down to: the trait bound `polkadot_test_runtime::RuntimeApiImpl<...>`: `polkadot_primitives::v1::ParachainHost<...>` is not satisfied This is WIP because it appears to be causing some std leakage into the wasm environment, or something; the compiler is currently complaining about duplicate definitions of `panic_handler` and `oom`. Presumably I have to identify all std types (Vec etc) and replace them with sp_std equivalents.
it wasn't std leakage, after all
* master: Bump Nominator Reward Limits (#1668)
Note: something changed in the `sc_cli::init_logger` call relatively recently; I had to update it slightly for things to work.
ordian
left a comment
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.
LGTM
|
Was this also tested on Polkadot nodes? We can not merge a pr that touches such a sensible part of the code base without testing it on Polkadot. |
This reverts commit af14ea5.
|
Nothing problematic, but on the burnin I see that we're running 15 tasks/node named |
|
Yes, for now. Replacing those with real subsystems is next up. |
We still need to test it at least once on Polkadot. |
| client.info().best_number, | ||
| 10, | ||
| ); | ||
| assert_eq!(client.info().best_number, 10,); |
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.
was this done intentionally or is an automatic formatter being used?
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.
In this case rustfmt generated the change.
| &get_header(10), | ||
| &get_header(10), | ||
| ), | ||
| voting_rule.restrict_vote(&*client, &get_header(0), &get_header(10), &get_header(10),), |
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.
with the exception of 1-item tuples, ,) is an invalid string in our codebase.
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.
OK: 372f4f6.
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.
Though I suspect that it is also valid when used appropriately in macros, i.e.
| ( $( $hash:expr ),* $(,)? ) => [ |
| = service::new_full_nongeneric( | ||
| config, None, authority_discovery_enabled, grandpa_pause, false, | ||
| = service::build_full( | ||
| config, None, authority_discovery_enabled, grandpa_pause, |
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.
invalid formatting (though this PR makes it no worse).
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.
OK: 944b745.
| target_header = backend.header(BlockId::Hash(target_hash)).ok()? | ||
| .expect("Header known to exist due to the existence of one of its descendents; qed"); | ||
| target_header = backend.header(BlockId::Hash(target_hash)).ok()?.expect( | ||
| "Header known to exist due to the existence of one of its descendents; qed", |
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.
previous formatting was better.
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.
OK: b7b45b3.
* Restore "Remove service, migrate all to service-new (#1630)" i.e. Revert "Revert "Remove service, migrate all to service-new (#1630)" (#1731)" This reverts commit c68aee3. This allows us to get the changeset from #1630 into a new branch which can be merged sometime in the future after appropriate burnin tests have completed. * remove ',)' from codebase outside of macros * restore bdfl-preferred formatting * attempt to improve destructuring formatting * rename polkadot-service-new -> polkadot-service * Apply suggestions from code review Co-authored-by: Bastian Köcher <[email protected]> * remove unused import * Update runtime/rococo-v1/README.md Co-authored-by: Bastian Köcher <[email protected]> Co-authored-by: Andronik Ordian <[email protected]> Co-authored-by: Bastian Köcher <[email protected]>
Makes progress towards #1620. Does not close that issue.
Integrates #1697.
Note: because this work involves replacing the old
servicewithservice-new, this needs some burn-in to ensure that we can still follow the existing chains and author blocks.Note: initially, the goal was to do all the work necessary for #1620 in this PR, hence the branch name. However, recently I've been spending a lot of time just keeping up with changes to
masterwhich updateservicebut notservice-new, and that gets old fast. There are therefore a couple of changes in here not related to the service update, which would probably be more effort than they're worth to extract.[edit] Burnin running on
kusama-sentry-ue1-0/kusama-validator-ue1-0