From 4ee661202e7199e9f254983b953098ae0e58d4e4 Mon Sep 17 00:00:00 2001
From: HHoflittlefish777 <77738092+HHoflittlefish777@users.noreply.github.com>
Date: Thu, 21 Dec 2023 14:01:05 +0800
Subject: [PATCH] [improve](transaction) extend abort transaction time (#28662)

---
 docs/en/docs/admin-manual/config/fe-config.md          | 10 ++++++++++
 docs/zh-CN/docs/admin-manual/config/fe-config.md       | 10 ++++++++++
 .../src/main/java/org/apache/doris/common/Config.java  |  7 +++++++
 .../java/org/apache/doris/system/HeartbeatMgr.java     |  3 ++-
 4 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/docs/en/docs/admin-manual/config/fe-config.md b/docs/en/docs/admin-manual/config/fe-config.md
index 2fd5b58d9f..b408a8156f 100644
--- a/docs/en/docs/admin-manual/config/fe-config.md
+++ b/docs/en/docs/admin-manual/config/fe-config.md
@@ -587,6 +587,16 @@ Is it possible to configure dynamically: true
 
 Whether it is a configuration item unique to the Master FE node: true
 
+### `abort_txn_after_lost_heartbeat_time_second`
+
+Abort transaction time after lost heartbeat. The default value is 300, which means transactions of be will be aborted after lost heartbeat 300s.
+
+Default: 300(s)
+
+Is it possible to configure dynamically: true
+
+Whether it is a configuration item unique to the Master FE node: true
+
 #### `enable_access_file_without_broker`
 
 Default：false
diff --git a/docs/zh-CN/docs/admin-manual/config/fe-config.md b/docs/zh-CN/docs/admin-manual/config/fe-config.md
index 47bcc9fba9..cdc8295557 100644
--- a/docs/zh-CN/docs/admin-manual/config/fe-config.md
+++ b/docs/zh-CN/docs/admin-manual/config/fe-config.md
@@ -587,6 +587,16 @@ FE向BE的BackendService发送rpc请求时的超时时间，单位：毫秒。
 
 是否为 Master FE 节点独有的配置项：true
 
+#### `abort_txn_after_lost_heartbeat_time_second`
+
+丢失be心跳后丢弃be事务的时间。默认时间为三百秒，当三百秒fe没有接收到be心跳时，会丢弃该be的所有事务。
+
+默认值：300(秒)
+
+是否可以动态配置：true
+
+是否为 Master FE 节点独有的配置项：true
+
 #### `enable_access_file_without_broker`
 
 默认值：false
diff --git a/fe/fe-common/src/main/java/org/apache/doris/common/Config.java b/fe/fe-common/src/main/java/org/apache/doris/common/Config.java
index 8c90f6a516..962bf219e4 100644
--- a/fe/fe-common/src/main/java/org/apache/doris/common/Config.java
+++ b/fe/fe-common/src/main/java/org/apache/doris/common/Config.java
@@ -1782,6 +1782,13 @@ public class Config extends ConfigBase {
     @ConfField(mutable = true, masterOnly = true)
     public static long max_backend_heartbeat_failure_tolerance_count = 1;
 
+    /**
+     * Abort transaction time after lost heartbeat.
+     * The default value is 300s, which means transactions of be will be aborted after lost heartbeat 300s.
+     */
+    @ConfField(mutable = true, masterOnly = true)
+    public static int abort_txn_after_lost_heartbeat_time_second = 300;
+
     /**
      * Heartbeat interval in seconds.
      * Default is 10, which means every 10 seconds, the master will send a heartbeat to all backends.
diff --git a/fe/fe-core/src/main/java/org/apache/doris/system/HeartbeatMgr.java b/fe/fe-core/src/main/java/org/apache/doris/system/HeartbeatMgr.java
index 4282ad9aa3..5d17846476 100644
--- a/fe/fe-core/src/main/java/org/apache/doris/system/HeartbeatMgr.java
+++ b/fe/fe-core/src/main/java/org/apache/doris/system/HeartbeatMgr.java
@@ -174,7 +174,8 @@ public class HeartbeatMgr extends MasterDaemon {
                     if (hbResponse.getStatus() != HbStatus.OK) {
                         // invalid all connections cached in ClientPool
                         ClientPool.backendPool.clearPool(new TNetworkAddress(be.getHost(), be.getBePort()));
-                        if (!isReplay && System.currentTimeMillis() - be.getLastUpdateMs() > 60 * 1000L) {
+                        if (!isReplay && System.currentTimeMillis() - be.getLastUpdateMs()
+                                >= Config.abort_txn_after_lost_heartbeat_time_second * 1000L) {
                             Env.getCurrentGlobalTransactionMgr()
                                     .abortTxnWhenCoordinateBeDown(be.getHost(), 100);
                         }