Conversation
|
e2e tests are successful, so I think we can ignore other failed as flaky |
@Wondertan Can you briefly explain how the src field caused |
|
@liamsi, yeah The test checks that samples are always not equal. However, the test was still failing, as sometimes equal samples were generated, due to |
Oh I think see: https://github.com/lazyledger/lazyledger-core/blob/50f722a510dd2ba8e3d31931c9d83132d6318d4b/p2p/ipld/sample.go#L61 the map uses the |
* Add changelog entry for #314 Signed-off-by: Thane Thomson <connect@thanethomson.com> * Build changelog with unreleased changes for pre-release Signed-off-by: Thane Thomson <connect@thanethomson.com> * version: Set to v0.34.27-alpha.1 Signed-off-by: Thane Thomson <connect@thanethomson.com> * ci: Update to incorporate/backport changes from #317 Signed-off-by: Thane Thomson <connect@thanethomson.com> * ci: Simplify release workflow further Signed-off-by: Thane Thomson <connect@thanethomson.com> * ci: Add step descriptions for pre-release and release workflows to explain what they do Signed-off-by: Thane Thomson <connect@thanethomson.com> --------- Signed-off-by: Thane Thomson <connect@thanethomson.com>
Originally Sample stored
srcfield not to sharemath/randobject. After migration tocrypto/randthat became redundant, though kept within the code, as previously unnoticed. The code also had the undesired side effects that causedTestSampleSquareto randomly fail. This PR fixes both issues.