You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@rocketmq.apache.org by ji...@apache.org on 2022/05/01 08:17:10 UTC
[rocketmq] branch 5.0.0-beta updated (83de6d305 -> d6a53f9c2)
This is an automated email from the ASF dual-hosted git repository.
jinrongtong pushed a change to branch 5.0.0-beta
in repository https://gitbox.apache.org/repos/asf/rocketmq.git
omit 83de6d305 Polish the document of SlaveActingMasterMode.md and QuorumACK.md
omit 48e2e2d4b Fix check style to paas CI
new ac886bc18 Fix check style to paas CI
new d6a53f9c2 Polish the document of SlaveActingMasterMode.md and QuorumACK.md
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (83de6d305)
\
N -- N -- N refs/heads/5.0.0-beta (d6a53f9c2)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
The 2 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
[rocketmq] 02/02: Polish the document of SlaveActingMasterMode.md and QuorumACK.md
Posted by ji...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
jinrongtong pushed a commit to branch 5.0.0-beta
in repository https://gitbox.apache.org/repos/asf/rocketmq.git
commit d6a53f9c29f30824b4c35a099e1caeaae63a6685
Author: RongtongJin <ji...@mails.ucas.ac.cn>
AuthorDate: Sun May 1 15:42:11 2022 +0800
Polish the document of SlaveActingMasterMode.md and QuorumACK.md
---
docs/cn/QuorumACK.md | 6 +++---
docs/cn/SlaveActingMasterMode.md | 3 +++
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/docs/cn/QuorumACK.md b/docs/cn/QuorumACK.md
index be609c6f8..e742f507d 100644
--- a/docs/cn/QuorumACK.md
+++ b/docs/cn/QuorumACK.md
@@ -34,11 +34,11 @@
通过Nameserver的反向通知以及GetBrokerMemberGroup请求可以获取当前副本组的存活信息，而Master与Slave的Commitlog高度差也可以通过HA服务中的位点记录计算出来。将增加以下参数完成自动降级：
-- **minInSyncReplicas**：最小需保持同步的副本组数量，仅在enableAutoInSyncReplicas为true时生效，默认为1。
+- **minInSyncReplicas**：最小需保持同步的副本组数量，仅在enableAutoInSyncReplicas为true时生效，默认为1
- **enableAutoInSyncReplicas**：自动同步降级开关，开启后，若当前副本组处于同步状态的broker数量（包括master自身）不满足inSyncReplicas指定的数量，则按照minInSyncReplicas进行同步。同步状态判断条件为：slave commitLog落后master长度不超过haSlaveFallBehindMax。默认为false。
-- **haSlaveFallBehindMax**：slave是否与master处于in-sync状态的判断值，slave commitLog落后master长度超过该值则认为slave已处于非同步状态。当enableAutoInSyncReplicas打开时，该值越小，越容易触发master的自动降级，当enableAutoInSyncReplicas关闭，且totalReplicas==inSyncReplicas时，该值越小，越容易导致在大流量时发送请求失败，故在该情况下可适当调大haSlaveFallBehindMax。默认为256K。
+- **haMaxGapNotInSync**：slave是否与master处于in-sync状态的判断值，slave commitLog落后master长度超过该值则认为slave已处于非同步状态。当enableAutoInSyncReplicas打开时，该值越小，越容易触发master的自动降级，当enableAutoInSyncReplicas关闭，且totalReplicas==inSyncReplicas时，该值越小，越容易导致在大流量时发送请求失败，故在该情况下可适当调大haMaxGapNotInSync。默认为256K。
-注意：在RocketMQ 4.x中存在haSlaveFallbehindMax参数，默认256MB，表明Slave与Master的CommitLog高度差多少后判定其为不可用，在RocketMQ 5中该参数被取消，由haSlaveFallBehindMax代替，含义如上。
+注意：在RocketMQ 4.x中存在haSlaveFallbehindMax参数，默认256MB，表明Slave与Master的CommitLog高度差多少后判定其为不可用，在RIP-34中该参数被取消。
```java
//计算needAckNums
diff --git a/docs/cn/SlaveActingMasterMode.md b/docs/cn/SlaveActingMasterMode.md
index 03978bfd3..7af7ff89f 100644
--- a/docs/cn/SlaveActingMasterMode.md
+++ b/docs/cn/SlaveActingMasterMode.md
@@ -134,9 +134,12 @@ Slave代理Master时，ScheduleMessageService将启动，时间到期的延迟
## 配置更新
Nameserver
+
- scanNotActiveBrokerInterval：扫描不活跃broker间隔，每次扫描将判断broker心跳是否超时，默认5s。
- supportActingMaster：nameserver端是否支持Slave代理Master模式，开启后，副本组在无master状态下，brokerId==1的slave将在TopicRoute中被替换成master（即brokerId=0），并以只读模式对客户端提供服务，默认为false。
+
Broker
+
- enableSlaveActingMaster：broker端开启slave代理master模式总开关，默认为false。
- enableRemoteEscape：是否允许远程逃逸，默认为false。
- brokerHeartbeatInterval：broker向nameserver发送心跳间隔（不同于注册间隔），默认1s。
[rocketmq] 01/02: Fix check style to paas CI
Posted by ji...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
jinrongtong pushed a commit to branch 5.0.0-beta
in repository https://gitbox.apache.org/repos/asf/rocketmq.git
commit ac886bc1820d76b057beb27cba2b3107105b8748
Author: RongtongJin <ji...@mails.ucas.ac.cn>
AuthorDate: Fri Apr 29 16:51:42 2022 +0800
Fix check style to paas CI
---
.../rocketmq/apis/consumer/MessageListener.java | 38 ++++++++++------------
1 file changed, 18 insertions(+), 20 deletions(-)
diff --git a/apis/src/main/java/org/apache/rocketmq/apis/consumer/MessageListener.java b/apis/src/main/java/org/apache/rocketmq/apis/consumer/MessageListener.java
index bcfab9748..a56e67781 100644
--- a/apis/src/main/java/org/apache/rocketmq/apis/consumer/MessageListener.java
+++ b/apis/src/main/java/org/apache/rocketmq/apis/consumer/MessageListener.java
@@ -24,33 +24,31 @@ import org.apache.rocketmq.apis.message.MessageView;
* synchronously.
*
* <p>PushConsumer will fetch messages from brokers and dispatch them to an
- * embedded thread pool in form of <code>Runnable</code> tasks to achieve
- * desirable processing concurrency.
+ * embedded thread pool in form of <code>Runnable</code> tasks to achieve desirable processing concurrency.
*
* <p>Refer to {@link PushConsumer} for more further specs.
*
* <p>
- * <strong>Thread Safety</strong>
- * This class may be called concurrently by multiple threads. Implementation should be thread safe.
+ * <strong>Thread Safety</strong>
+ * This class may be called concurrently by multiple threads. Implementation should be thread safe.
* </p>
*/
public interface MessageListener {
- /**
- * Callback interface to handle incoming messages.
- *
- * Application developers are expected to implement this interface to fulfill
- * business requirements through processing <code>message</code> and return
- * <code>ConsumeResult</code> accordingly.
- *
- * PushConsumer will, on behalf of its group, acknowledge the message to
- * broker on success; In case of failure or unexpected exceptions
- * were raised, it will negatively acknowledge <code>message</code>, which
- * would potentially get re-delivered after the configured back off period.
- *
- * @param message The message passed to the listener.
- * @return {@link ConsumeResult#SUCCESS} if <code>message</code> is properly processed; {@link ConsumeResult#FAILURE}
- * otherwise.
- */
+ /**
+ * Callback interface to handle incoming messages.
+ *
+ * Application developers are expected to implement this interface to fulfill business requirements through
+ * processing <code>message</code> and return
+ * <code>ConsumeResult</code> accordingly.
+ *
+ * PushConsumer will, on behalf of its group, acknowledge the message to broker on success; In case of failure or
+ * unexpected exceptions were raised, it will negatively acknowledge <code>message</code>, which would potentially
+ * get re-delivered after the configured back off period.
+ *
+ * @param message The message passed to the listener.
+ * @return {@link ConsumeResult#SUCCESS} if <code>message</code> is properly processed; {@link
+ * ConsumeResult#FAILURE} otherwise.
+ */
ConsumeResult onMessage(MessageView message);
}