Skip to content
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

v0.26.3 merge master #2541

Merged
merged 270 commits into from
Jun 2, 2022
Merged

v0.26.3 merge master #2541

merged 270 commits into from
Jun 2, 2022

Conversation

peterargue and others added 30 commits May 9, 2022 20:45
…r.go

Co-authored-by: Leo Zhang <zhangchiqing@gmail.com>
…r.go

Co-authored-by: Leo Zhang <zhangchiqing@gmail.com>
lolpuddle and others added 19 commits June 1, 2022 15:35
1895: [State Sync] Add execution data requester r=peterargue a=peterargue

Closes #1452
Closes #1897

This PR adds a new component for requesting `ExecutionData` from the network for newly sealed blocks. It does so by watching for finalized block notifications from hotstuff, and requesting data for each of the blocks included in the seals.

This currently sets no expiration for the state data, which means nodes will keep the data for the duration of the spork. We're expecting that to be ~150GB per month. Theoretically, an operator could delete the execution data db, and restart the node with a specific start block if they need/want to prune. Longer term, we'll want to provide better options for managing disk usage.

To Complete:

- [x] Add to access node
- [x] Add metrics
- [x] Unit tests
- [x] Integration tests

Co-authored-by: Peter Argue <89119817+peterargue@users.noreply.github.com>
2514: Add BlockHeight to TransactionResultResponse r=zhangchiqing a=lolpuddle

Depends on updates to .proto files for [this PR](onflow/flow#952)

[Closes 2413](#2413)



Co-authored-by: lolpuddle <oocean.cheung@gmail.com>
2524: Exit validation early for empty collections r=jordanschalm a=jordanschalm

For empty collections, the reference height range was not set correctly, which resulted in checking for duplicates in all cluster blocks. This PR exits validation early for empty collections.

⚠️  This PR also adds a mechanism for irrecoverable HotStuff errors, which currently are silently dropped. An issue to implement a more robust solution to this problem is in https://github.com/dapperlabs/flow-go/issues/6263.

Notes: https://www.notion.so/dapperlabs/Transaction-De-Dupe-Deployment-Notes-8027bde5183a4e4c931bb1d42905952a

Co-authored-by: Jordan Schalm <jordan@dapperlabs.com>
2509: [Crypto] improve flaky random test r=tarakby a=tarakby

 A rare flakiness has been seen in a statistical tests under crypto/random 
 
 - use math rand and add logging for failing tests seeds
 - use higher sampling for a statistical test to avoid flakiness (the higher the sampling, the closer the result to theoretical distribution)
 
 cc `@durkmurder` 

Co-authored-by: Tarak Ben Youssef <tarak.benyoussef@dapperlabs.com>
Co-authored-by: Tarak Ben Youssef <50252200+tarakby@users.noreply.github.com>
2167: Add secure-cadence prefix as a tag to be pushed r=Kay-Zee a=Kay-Zee

Previously, our makefile only recognizes tags starting with `v` as acceptable, this just adds the `secure-cadence` prefix as one that we're also building

Co-authored-by: Kay-Zee <kan@axiomzen.co>
Co-authored-by: Kan Zhang <kan@axiomzen.co>
2511: chore(tests): do not go through the resolver for localhost r=SaveTheRbtz a=SaveTheRbtz

Currently, we go through the resolver codepath each time we create a local connection.  This is not too expensive for the benchmark (though it can create unexpected spikes in latencies), but it can be expensive for the docker's resolver.  To avoid DDoS'ing it, let's use `127.0.0.1` for now.  If our nodes switch to IPv6-only we can switch this to `::1`.

Co-authored-by: Alexey Ivanov <SaveTheRbtz@GMail.com>
2519: chore(tests): remove unused batchLoadGenerator r=SaveTheRbtz a=SaveTheRbtz

This has been unused for quite a while and replaced with a contLoadGenerator

Co-authored-by: Alexey Ivanov <SaveTheRbtz@GMail.com>
PR Migrate ingestion engine to new Component interface
[Fix] add explicit mapping of cluster prefix -> message codes
@zhangchiqing zhangchiqing changed the base branch from master to v0.26 June 2, 2022 21:02
@zhangchiqing zhangchiqing marked this pull request as ready for review June 2, 2022 21:04
@zhangchiqing zhangchiqing merged commit c488689 into v0.26 Jun 2, 2022
@zhangchiqing zhangchiqing deleted the leo/v0.26.3-merge-master branch June 2, 2022 21:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.