Zookeeper is the child project of apache hadoop. Since it offers tree-like directory service and supports change notification, it’s suitable to use it as dubbo’s registry server. It’s a field-proven product, therefore it’s recommended to use it in the production environment. 1
Description on registration procedure:
/dubbo/com.foo.BarService/providersfor provider’s URL addresses. At the same time, write consumer’s URL address under
/dubbo/com.foo.BarServicefor the URL addresses from all providers and consumers.
The following abilities are supported:
<dubbo:registry check="false" />is configured, failed requests for subscription and registration will be recorded and kept retrying in the background.
<dubbo:registry username="admin" password="1234" />for zookeeper login.
<dubbo:registry group="dubbo" />for dubbo’s root node on zookeeper. Default root node will be used if it’s not specified.
<dubbo:reference group="*" version="*" />in order to subscribe all groups and all versions for the services to be referenced.
Add zookeeper client dependency in both provider and consumer:
<dependency> <groupId>org.apache.zookeeper</groupId> <artifactId>zookeeper</artifactId> <version>3.3.3</version> </dependency>
Or download directly from apache.
Dubbo supports two zookeeper clients: zkclient and curator:
2.2.0 dubbo uses zkclient by default, in order to improve the robustness. zkclient is a zookeeper client implementation open-sourced by Datameer.
Note: The implementation of zkclient has been removed in the 2.7.x version. If you want to use the zkclient client, you need to extend it yourself.
<dubbo:registry ... client="zkclient" />
In order to use it, need to explicitly declare the following maven dependency or download its client.
<dependency> <groupId>com.github.sgroschupf</groupId> <artifactId>zkclient</artifactId> <version>0.1</version> </dependency>
2.3.0 dubbo also supports curator but explicit configuration is required. Curator is the zookeeper client open-sourced by Netflix.
In order to switch to curator, use the configuration below:
<dubbo:registry ... client="curator" />
Also need to explicitly add maven dependency or directly download the jar:
<dependency> <groupId>com.netflix.curator</groupId> <artifactId>curator-framework</artifactId> <version>1.1.10</version> </dependency>
Zookeeper single node configuration:
<dubbo:registry address="zookeeper://10.20.153.10:2181" />
<dubbo:registry protocol="zookeeper" address="10.20.153.10:2181" />
Zookeeper cluster configuration：
<dubbo:registry address="zookeeper://10.20.153.10:2181?backup=10.20.153.11:2181,10.20.153.12:2181" />
<dubbo:registry protocol="zookeeper" address="10.20.153.10:2181,10.20.153.11:2181,10.20.153.12:2181" />
Configure single zookeeper to serve as multiple registry servers:
<dubbo:registry id="chinaRegistry" protocol="zookeeper" address="10.20.153.10:2181" group="china" /> <dubbo:registry id="intlRegistry" protocol="zookeeper" address="10.20.153.10:2181" group="intl" />
Pls. refer to zookeeper install manual for how to install zookeeper based registry server. To set it up, specify
conf/dubbo.properties for both provider and consumer (you can refer to quick start) after install a zookeeper server.
A home-brewed service registry server is used in Alibaba instead of zookeeper server. Zookeeper based registry center does not have long-run practice within Alibaba, therefore we cannot guarantee its reliability. Zookeeper registry server is provided for dubbo community, and its reliability relies on zookeeper itself largely.
The original designed data structure for zookeeper in
2.0.8 has the limitation that data type cannot extended, it’s redesigned in
2.0.9. But at the same time incompatibility is introduced, thereby
2.0.9 is required for all service providers and service consumers.
2.2.0 zkclient is used by default, therefore its dependency is needed.
2.3.0 curator is supported as alternative option.
Suggest to use
2.3.3 or above for zookeeper registry client ↩︎
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.