Shijie Sheng
Shijie Sheng
**What changed?** Release note for v0.20.0 **Why?** **How did you test it?** **Potential risks**
**What changed?** * moved peerResolver to its own package * refactor history client to uniformly use peerResolver's `FromRedirectKey` method * [planned] use codegen to generate redirect wrapper **Why?** simplify history...
**Why** Bug is introduced in https://github.com/uber/cadence-java-client/pull/876 **Symptom** Memory leak happens on every poller when tracing is enabled. Span objects grows since scope is not closed correctly. **Mitigation** Disable the tracer...
**What changed?** **Why?** **How did you test it?** **Potential risks** **Release notes** **Documentation Changes**
**Describe the bug** Decision worker will stop processing if decision poller count is 1 and sticky tasklist is enabled **To Reproduce** MaxConcurrentDecisionTaskPollers is set to 1 (default 2) Or MinConcurrentDecisionTaskPollers...
**What changed?** * Populate TaskListKind for sticky request **Why?** Without the correct king, it might remove TTL on sticky tasklist records **How did you test it?** **Potential risks**
**Why?** Fanout pattern is common in user workflows. However, it's only possible to fanout **How did you test it?** Unit Test in test environment **Detailed Description** * Added `NewBatchFuture` API...
**Detailed Description** Improve performance of poller auto scaler by using more accurate scaling signals and several implementation changes. Changes * New WorkerOptions `AutoScalerOptions` is introduced. * Several WorkerOptions are deprecated...
**What changed?** * remove thrift encoding/decoding support as default in Data Converter **Why?** Thrift deprecation **How did you test it?** Unit Test. Internally, we no longer have users depend on...
**What changed?** **Why?** **How did you test it?** **Potential risks** **Release notes** **Documentation Changes**