- Add recent argument parameter for transformNullFunction #75
- Improved tag autocomplete request performance. #69
- Users have to migrate documents in Elasticsearch with script located in scripts/v1.9.1/es-index-migration.py
- Support Prometheus Lexer. #62
- Support basic auth for Elasticsearch client #64
- Drop metric if metric id is greater than 512 bytes. It is es key hard limit. #64
- Improve local key cache #60
- Include all option on tag value whenever result set is not empty #59
- Configurable resolution(rollup) #63
- Read and write metrics from Cassandra by following offset based scheme #44
- Should deserialize converted metric if label value has whitespace #54
- Add input kafka type with graphite deserializer #43
- Search Graphene metrics by tags #38
- Add prometheus deserializer in kafka input. #48
- Extract elasticsearch port and protocol to property #42 Contributed by @jacek-bakowski
- Use index property each handler type #51
- Increase graphene cassandra handler max queue size #40
- Fix several bugs #46
- Fix prometheus deserializer edge case #51
- Add Errors enum for handling exception more easily #37
- Fix rotated index with leading zero #34
- add index-based key search handler & query optimizer for performance #32
- Add KeyCache interface for checking duplication metric key #25
- Add helm charts for support easy installation on k8s #28 Contributed by @hwanjin-jeong
- Add nginx docker image for graphene reader's proxy #27
- Add dynamic template to store each key's parts in keyword type #30
- Add bulkAsync feature for indexing performance #20
- Change package name to com.graphene #31
- Bump up dependency version #29
- Ignore unavailable index when read. #21
- Improve Key Store throughput #23
- Prevent calling clearScroll if there's no scrollId. #24
- Fixed some bugs and removed unnecessary codes and operations. #26
- Support past metric indexing based on metric timestamp. #18
- Full support for MapSeries, ReduceSeries functions. #19
- Index is not determined by metric's timestamp when writing. #17
- Modify bulk interval unit to milliseconds
- Bump up elasticsearch client version to 6.8.1
- Add auto_expand_replicas property
- Reduce unnecessary copy operation when using StringBuilder to make response. #5
- Support multiple handlers using property. #6
- Add the index based key store handler for performance. #7
- The Controller just uses the content set in FullHttpResponse and change it to String AGAIN to make response body to be handled by Spring's RestController, which are all unnecessary. #10
- Support elasticsearch 6.x.x with index template and create automatically index. #11
- Add index rotating feature with the tenant ( daily, weekly ). #13
- Support rotating index query. #14
- Add multi-key store handler. #15
- If time-series data is empty, Graphene returns ']' single string. #8
- Try entire code refactoring to apply index based key store handler
- Massive performance and stability improvements
- aliasByNode function now supports minus index #2
- Fix metrics/find API duplicated path issue #3
- Remove local aggregation and rollup feature
- Development environment to use docker-compose.yml
- Support the /metrics/find API to use at Grafana #1
- Implement StatsService interface to be disable graphite stats service
- Change project build tool from maven to gradle
- Use spring framework for graphene