Zhangao
Zhangao
fix #472
`pulsar.reader.use-earliest-when-data-loss` property has been transformed into `useEarliestWhenDataLoss` by method `getPulsarProperties` in `readerConf`. So methdo `getUseEarliestWhenDataLossAndRemoveKey` should get `useEarliestWhenDataLoss` from readerConf instead of `use-earliest-when-data-loss`.
Nowadays, sql doesn't support pulsar client configuration. Like, `memoryLimitBytes`
Reproduce step: 1. Consumer a topic using kafka clients. 2. delete topic using `bin/pulsar-admin topics delete-partitioned-topic` 3. Get consumer group using `bin/kafka-consumer-groups.sh --list`
KOP 2.8.1.23 Pulsar 2.8 kafka client 2.4.1 Message produced failed: (other topics produced successfully, but some got the error. ) ``` 2021-11-26 10:34:19.061 [kafka-producer-network-thread | producer-1] WARN org.apache.kafka.clients.producer.internals.Sender Sender.java:637 -...
**Bookkeeper Version: 4.14.4** Our prod environment went wrong last week, all bookies were killed because of direct memory OOM, this happened after one bookie's disk was broken, we tried to...
### Motivation SyncThread flush would invoke `checkpointSource.checkpointComplete` twice. it's unnecessary. ### Changes Change `org.apache.bookkeeper.bookie.LedgerStorage#flush` signature . add `doCheckpointComplete` parameter. default value is true, only set to false in `org.apache.bookkeeper.bookie.SyncThread#flush`
### Motivation Admin command `stats-internal` support hiding ledgers list. If there are thousands of ledgers, it's meaningless to print so much ledgers. In additional, `partitioned-stats-internal` also support this feature. ###...
### Search before asking - [X] I searched in the [issues](https://github.com/apache/pulsar/issues) and found nothing similar. ### Version Master ### Minimal reproduce step just run bin/pulsar-admin ### What did you expect...