Skip to content

Commit 7cd1d66

Browse files
author
Cary Huang
committed
v1.2
1 parent 7955065 commit 7cd1d66

File tree

4 files changed

+100
-2
lines changed

4 files changed

+100
-2
lines changed

docs/en/changelog.md

Lines changed: 51 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -4,6 +4,57 @@ All notable changes to this project will be documented in this file.
44
The format is based on [Keep a Changelog](http://keepachangelog.com/)
55
and this project adheres to [Semantic Versioning](http://semver.org/).
66

7+
## **[SynchDB 1.2](https://github.com/Hornetlabs/synchdb/releases/tag/v1.2) - 2025-09-03**
8+
9+
SynchDB 1.2 introduces introduces a native Openlog Replicator connector (BETA), enhanced monitoring with JMX and Grafana, and a new ezdeploy.sh tool for quick deployments and testing. It also adds snapshot table selection, performance improvements, and key fixes, while addressing connector isolation and stability issues.
10+
11+
### **Added**
12+
13+
#### [Native Openlog Replicator Connector](https://docs.synchdb.com/user-guide/configure_olr/) - BETA
14+
* Added `synchdb_add_olr_conninfo` and `synchdb_del_olr_conninfo` to enable or disable openlog replicator based streaming.
15+
* Added a new connector type `olr`, which is a native (no Debezium) openlog replicator client that stream Oracle database changes from an external Openlog Replicator service.
16+
* Supported DMLs: insert, update delete
17+
* Supported DDLs: CREATE TABLE, DROP TABLE, ALTER TABLE MODIFY, ALTER TABLE ADD/DROP COLUMN, ALTER TABLE ADD/DROP CONSTRAINT, TRUNCATE
18+
* Based on libprotobuf-c to communiate with Openlog Replicator and IvorySQL's Oracle parser to process incoming DDL query events,
19+
* Supported snapshot modes: initial, initial_only, no_data, always, never
20+
* Supported batching, schema history, and offset management just like other Debezium based connectors.
21+
* Supported Debezium-based Openlog Replicator connector in addition to native.
22+
23+
#### [Monitoring](https://docs.synchdb.com/monitoring/jmx_monitor/)
24+
25+
* Added `synchdb_add_jmx_conninfo` and `synchdb_del_jmx_conninfo` to enable or disable JMX based monitoring.
26+
* Added `synchdb_add_jmx_exporter_conninfo` and `synchdb_del_jmx_exporter_conninfo` to enable and disable monitoring support with Prometheus and Grafana.
27+
* Added Grafana based dashboard templates for supported Debezium based connectors (MySQL, SQL Server and Oracle).
28+
29+
#### [ezdeploy.sh](https://docs.synchdb.com/getting-started/quick_start/)
30+
* Added `ezdeploy.sh` tool that can quickly deploy a pre-build SynchDB and selected source database types to do quick connector testing.
31+
* supported deployment: MySQL, SQL Server, Oracle23ai, Oracle19c, Openlog Replicator 1.3.0
32+
* supported Prometheus and Grafana deployment with preloaded dashboards.
33+
* Supported pre-compiled SynchDB v1.2 for quick deployment + tests
34+
35+
36+
### **Changed**
37+
38+
* Added a new argument called `snapshot table` in `synchdb_add_conninfo` to allow users to select which tables to redo initial snapshot when started in `always` snapshot mode.
39+
* Updated pytest framework to support hammerdb based TPC tests for Oracle.
40+
* Enhanced the performance of event polling between Debezium engine and SynchDB by using direct buffer instead of frequent JNI calls.
41+
* When a connector is resumed via `synchdb_resume_engine`, it will always resume in `initial` snapshot mode rather than the mode it was first started with.
42+
* `synchdb_state_view` and `synchdb_stats_view` will now only display connector information created by the current SynchDB extension. Connectors created by other SynchDB extensions in different databases will not be shown.
43+
44+
45+
### **Fixed**
46+
47+
* Fixed a crash in `spi_execute_select_one()` where it would return a reference that has been destroyed by SPI memory context at the end of a successful SPI execution.
48+
* Resolved compilation issues with SynchDB when PostgreSQL is built with cassert.
49+
* Fixed an issue where multiple connectors created with the same name by multiple SynchDB extensions (created in different databases) overwrite each other's shared memory data.
50+
51+
### **Known Issues and Additional Info**
52+
53+
* Native Openlog Replicator Connector currently stream all tables under specified database. Table Filtering is to be configured in Openlog Replicator rather than SynchDB.
54+
* Prometheus and Grafana based monitoring require JMX Exporter which can be downloaded [here](https://github.com/prometheus/jmx_exporter)
55+
56+
57+
758
## **[SynchDB 1.1](https://github.com/Hornetlabs/synchdb/releases/tag/v1.1) - 2025-04-17**
859

960
SynchDB 1.1 introduces Oracle connector support, enhanced data type transformation capabilities, and significantly improved core data processing engine. This update enhances performance through intelligent caching and optimized JSON parsing, while extending compatibility with PostgreSQL 16, 17, and IvorySQL 4.4.

docs/en/getting-started/quick_start.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -292,7 +292,7 @@ postgres=# SELECT * FROM synchdb_state_view;
292292
mysqlconn | mysql | 579845 | initial snapshot | polling | no error | {"ts_sec":1741301103,"file":"mysql-bin.000009","pos":574318212,"row":1,"server_id":223344,"event":2}
293293
oracleconn | oracle | 580053 | initial snapshot | polling | no error | offset file not flushed yet
294294
ora19cconn | oracle | 593421 | initial snapshot | polling | no error | offset file not flushed yet
295-
olrconn | oracle | 601235 | schema sync | polling | no error | offset file not flushed yet
295+
olrconn | oracle | 601235 | initial snapshot | polling | no error | offset file not flushed yet
296296
(5 rows)
297297
298298
```

docs/zh/changelog.md

Lines changed: 47 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -4,6 +4,53 @@
44
本文格式基于 [Keep a Changelog](http://keepachangelog.com/)
55
且本项目遵循 [语义化版本](http://semver.org/)
66

7+
## **[SynchDB 1.2](https://github.com/Hornetlabs/synchdb/releases/tag/v1.2) - 2025-09-03**
8+
9+
SynchDB 1.2 引入了原生 Openlog Replicator 连接器(测试版)、增强的 JMX 和 Grafana 监控功能,以及用于快速部署和测试的全新 ezdeploy.sh 工具。此外,它还增加了快照表选择功能、性能改进和关键修复,同时解决了连接器隔离和稳定性问题。
10+
11+
### **新增**
12+
13+
#### [原生 Openlog Replicator 连接器](https://docs.synchdb.com/user-guide/configure_olr/) - 测试版
14+
* 新增 `synchdb_add_olr_conninfo``synchdb_del_olr_conninfo`,用于启用或禁用基于 Openlog Replicator 的流式传输。
15+
* 添加了新的连接器类型“olr”,它是一个原生(非 Debezium)Openlog 复制器客户端,可从外部 Openlog Replicator 服务流式传输 Oracle 数据库更改。
16+
* 支持的 DML:插入、更新/删除
17+
* 支持的 DDL:创建表、删除表、修改表、添加/删除列、添加/删除约束、截断
18+
* 基于 libprotobuf-c 与 Openlog Replicator 和 IvorySQL 的 Oracle 解析器通信,以处理传入的 DDL 查询事件。
19+
* 支持的快照模式:initial、initial_only、no_data、always、never
20+
* 与其他基于 Debezium 的连接器一样,支持批处理、模式历史记录和偏移量管理。
21+
* 除原生连接器外,还支持基于 Debezium 的 Openlog Replicator 连接器。
22+
23+
#### [监控](https://docs.synchdb.com/monitoring/jmx_monitor/)
24+
25+
* 新增 `synchdb_add_jmx_conninfo``synchdb_del_jmx_conninfo`,用于启用或禁用基于 JMX 的监控。
26+
* 新增 `synchdb_add_jmx_exporter_conninfo``synchdb_del_jmx_exporter_conninfo`,用于启用或禁用 Prometheus 和 Grafana 的监控支持。
27+
* 新增基于 Grafana 的仪表板模板,用于支持基于 Debezium 的连接器(MySQL、SQL Server 和 Oracle)。
28+
29+
#### [ezdeploy.sh](https://docs.synchdb.com/getting-started/quick_start/)
30+
* 新增 `ezdeploy.sh` 工具,可快速部署预构建的 SynchDB 和所选的源数据库类型,以便快速进行连接器测试。
31+
* 支持部署:MySQL、SQL Server、Oracle23ai、Oracle19c、Openlog Replicator 1.3.0
32+
* 支持预加载仪表板的 Prometheus 和 Grafana 部署。
33+
* 支持预编译的 SynchDB v1.2,以便快速部署和测试。
34+
35+
### **变更**
36+
37+
*`synchdb_add_conninfo` 中添加了一个名为 `snapshot table` 的新参数,允许用户在以 `always` 快照模式启动时选择要重做初始快照的表。
38+
* 更新了 pytest 框架,以支持基于 hammerdb 的 Oracle TPC 测试。
39+
* 通过使用直接缓冲区代替频繁的 JNI 调用,增强了 Debezium 引擎和 SynchDB 之间事件轮询的性能。
40+
* 当连接器通过 `synchdb_resume_engine` 恢复时,它将始终以 `initial` 快照模式恢复,而不是首次启动时的模式。
41+
* `synchdb_state_view``synchdb_stats_view` 现在仅显示当前 SynchDB 扩展创建的连接器信息。其他 SynchDB 扩展在不同数据库中创建的连接器将不显示。
42+
43+
### **已修复**
44+
45+
* 修复了 `spi_execute_select_one()` 函数崩溃的问题,该函数在 SPI 执行成功后会返回一个已被 SPI 内存上下文销毁的引用。
46+
* 解决了使用 cassert 构建 PostgreSQL 时 SynchDB 的编译问题。
47+
* 修复了多个 SynchDB 扩展(创建于不同数据库中)创建的同名连接器会相互覆盖共享内存数据的问题。
48+
49+
### **已知问题及其他信息**
50+
51+
* 原生 Openlog Replicator 连接器目前会流式传输指定数据库下的所有表。表过滤功能需要在 Openlog Replicator 中配置,而不是在 SynchDB 中。
52+
* 基于 Prometheus 和 Grafana 的监控需要 JMX Exporter,可以从[此处](https://github.com/prometheus/jmx_exporter) 下载
53+
754
## **[SynchDB 1.1](https://github.com/Hornetlabs/synchdb/releases/tag/v1.1) - 2025-04-17**
855

956
SynchDB 1.1 版本引入了 Oracle 连接器支持,增强了数据类型转换能力,并显著改进了核心数据处理引擎。本次更新通过智能缓存和优化的 JSON 解析,提升了性能表现,同时扩展了与 PostgreSQL 16、17 和 IvorySQL 4.4 的兼容性。

docs/zh/getting-started/quick_start.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -293,7 +293,7 @@ postgres=# SELECT * FROM synchdb_state_view;
293293
mysqlconn | mysql | 579845 | initial snapshot | polling | no error | {"ts_sec":1741301103,"file":"mysql-bin.000009","pos":574318212,"row":1,"server_id":223344,"event":2}
294294
oracleconn | oracle | 580053 | initial snapshot | polling | no error | offset file not flushed yet
295295
ora19cconn | oracle | 593421 | initial snapshot | polling | no error | offset file not flushed yet
296-
olrconn | oracle | 601235 | schema sync | polling | no error | offset file not flushed yet
296+
olrconn | oracle | 601235 | initial snapshot | polling | no error | offset file not flushed yet
297297
(5 rows)
298298

299299
```

0 commit comments

Comments
 (0)