zhoujinsong
released this
20 Nov 11:20
·
146 commits
to master
since this release
Improvements:
- Support graceful shutdown for AMS #3057
- Prevent the table name from being truncated prematurely #3077
- Load log4j configuration file from AMORO_CONF_DIR #3098
- Make yarn application name for flink optimizer more meaningful #3129
- Filter out the CodecPool log in the optimizer log #3145
- Optimize plan log print #3212
- Do not throw exception when retry task in optimizer keeper #3110
- Add missing optimizingStatus key and format code #3152
- Fix wrong javadoc for TableRuntimeRefreshExecutor #3051
- Fix typo in managing-optimizers.md #3059
- Equality field id are different in a RewriteFilesInput #2870
- Make plugin package option #3302
- Optimize the final tarball to keep it within the specified size for 0.7 #3303
- Change amoro shade version to 0.7.0-incubating #3083
BugFixes:
- Fix AMS crashes if javalin stop timeouts #3052
- Fix the function that convert ms #3069
- Tag dangling delete snapshot commits #3058
- High available service should not curator client before LeaderLatch closing #3132
- When a master/backup switch occurs, close the planExecutor thread pool #3141
- Thrift port automatically closed #3153
- Optimization state blocking in planning #3111
- Optimization state blocking in running #3110
- Extract error message in optimizers #3130
- Fix scheduling hungry issue #3196
- OptimizerKeeper throws NPE if optimizer timeout exactly when ams restarted #3242
- Cannot set self-optimizing.min-plan-interval table property dynamically #3224
- Iceberg table ran repeated minor optimizing #3228
- Error in docker-compose.yaml of quick start #3086
- Planning table failed, java.lang.StackOverflowError: null #3239
- Got
java.lang.NoClassDefFoundError: org/apache/hadoop/shaded/com/ctc/wstx/io/InputBootstrapper
error when execute sql in terminal #3323