GTID环境中手动修复主从故障一例（Error 1236/Error 1396）
Preface
I got an replication error 1236 when I modified the password of a user without starting slave threads of replication.Further more,the user was absent in slave.below is the whole operation how I solve the issue.
Procedure
I modified the user password on master at first.
root@localhost:mysql3306.sock [(none)]>alter user zlm@'192.168.56.%' identified by 'zlmzlm';
Query OK, rows affected (0.00 sec)
I then started the slave replication on zlm3.
#mysql -uroot -p -S /var/lib/mysql/mysql.sock
Enter password:
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is
Server version: 5.7.-log MySQL Community Server (GPL) Copyright (c) , , Oracle and/or its affiliates. All rights reserved. Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners. Type 'help;' or '\h' for help. Type '\c' to clear the current input statement. root@localhost:mysql.sock [(none)]>start slave;
Query OK, rows affected (0.00 sec) root@localhost:mysql.sock [(none)]>show slave status\G
*************************** . row ***************************
Slave_IO_State:
Master_Host: 192.168.56.100
Master_User: repl
Master_Port:
Connect_Retry:
Master_Log_File: mysql-bin.
Read_Master_Log_Pos:
Relay_Log_File: relay-bin.
Relay_Log_Pos:
Relay_Master_Log_File: mysql-bin.
Slave_IO_Running: No //The IO_Thread is not running on account of Error 1236
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno:
Last_Error:
Skip_Counter:
Exec_Master_Log_Pos:
Relay_Log_Space:
Until_Condition: None
Until_Log_File:
Until_Log_Pos:
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno:
Last_IO_Error: Got fatal error from master when reading data from binary log: 'The slave is connecting using CHANGE MASTER TO MASTER_AUTO_POSITION = 1, but the master has purged binary logs containing GTIDs that the slave requires.'
Last_SQL_Errno:
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id:
Master_UUID: 2a4b3562-2ab6-11e8-be7a-080027de0e0e
Master_Info_File: /data/mysql/mysql3306/data/master.info
SQL_Delay:
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count:
Master_Bind:
Last_IO_Error_Timestamp: ::
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set:
Executed_Gtid_Set: 27af30ca--11e8-ad7e-080027de0e0e:-,
2a4b3562-2ab6-11e8-be7a-080027de0e0e:- //This is the id of maximum transaction executed on slave.
Auto_Position:
Replicate_Rewrite_DB:
Channel_Name:
Master_TLS_Version:
row in set (0.00 sec)
I decided to set gtid_purged variable to a certain value.So I executed "reset master;" to remove the value of "Executed_Gtid_Set".
root@localhost:mysql.sock [(none)]>reset master;
Query OK, rows affected (0.05 sec) root@localhost:mysql.sock [(none)]>show slave status\G
*************************** . row ***************************
Slave_IO_State:
Master_Host: 192.168.56.100
Master_User: repl
Master_Port:
Connect_Retry:
Master_Log_File: mysql-bin.
Read_Master_Log_Pos:
Relay_Log_File: relay-bin.
Relay_Log_Pos:
Relay_Master_Log_File: mysql-bin.
Slave_IO_Running: No
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno:
Last_Error:
Skip_Counter:
Exec_Master_Log_Pos:
Relay_Log_Space:
Until_Condition: None
Until_Log_File:
Until_Log_Pos:
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno:
Last_IO_Error: Got fatal error from master when reading data from binary log: 'The slave is connecting using CHANGE MASTER TO MASTER_AUTO_POSITION = 1, but the master has purged binary logs containing GTIDs that the slave requires.'
Last_SQL_Errno:
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id:
Master_UUID: 2a4b3562-2ab6-11e8-be7a-080027de0e0e
Master_Info_File: /data/mysql/mysql3306/data/master.info
SQL_Delay:
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count:
Master_Bind:
Last_IO_Error_Timestamp: ::
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set:
Executed_Gtid_Set: //The value has been cleared after operation of "reset master;".
Auto_Position:
Replicate_Rewrite_DB:
Channel_Name:
Master_TLS_Version:
row in set (0.00 sec)
In order to know which value of "gtid_purged" I should set,I analyzed the binlog on master.
zlm@192.168.56.100: [(none)]>show master status;
+------------------+----------+--------------+------------------+-------------------------------------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |
+------------------+----------+--------------+------------------+-------------------------------------------------+
| mysql-bin. | | | | 2a4b3562-2ab6-11e8-be7a-080027de0e0e:- |
+------------------+----------+--------------+------------------+-------------------------------------------------+
row in set (0.00 sec) zlm@192.168.56.100: [(none)]>flush logs;
Query OK, rows affected (0.04 sec) zlm@192.168.56.100: [(none)]>show master status;
+------------------+----------+--------------+------------------+-------------------------------------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |
+------------------+----------+--------------+------------------+-------------------------------------------------+
| mysql-bin. | | | | 2a4b3562-2ab6-11e8-be7a-080027de0e0e:- |
+------------------+----------+--------------+------------------+-------------------------------------------------+
row in set (0.00 sec) [root@zlm1 :: /data/mysql/mysql3306/logs]
#mysqlbinlog -vv --base64-output=decode-rows mysql-bin. > /root/.log
[root@zlm1 :: /data/mysql/mysql3306/logs]
#cd [root@zlm1 :: ~]
#ls -l|grep .log
-rw-r--r-- root root Jul : .log
[root@zlm1 :: ~]
#less .log
... /*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
# at
# :: server id end_log_pos CRC32 0x6e71f8f2 Start: binlog v , server v 5.7.-log created :: at startup
ROLLBACK/*!*/;
# at
# :: server id end_log_pos CRC32 0x074cfb08 Previous-GTIDs
# 2a4b3562-2ab6-11e8-be7a-080027de0e0e:-
# at
# :: server id end_log_pos CRC32 0x9125f364 GTID last_committed= sequence_number= rbr_only=no
SET @@SESSION.GTID_NEXT= '2a4b3562-2ab6-11e8-be7a-080027de0e0e:12715627'/*!*/;
# at
# :: server id end_log_pos CRC32 0x7a5353bf Query thread_id= exec_time= error_code=
SET TIMESTAMP=/*!*/;
SET @@session.pseudo_thread_id=/*!*/;
SET @@session.foreign_key_checks=, @@session.sql_auto_is_null=, @@session.unique_checks=, @@session.autocommit=/*!*/;
SET @@session.sql_mode=/*!*/;
SET @@session.auto_increment_increment=, @@session.auto_increment_offset=/*!*/;
/*!\C utf8 *//*!*/;
SET @@session.character_set_client=,@@session.collation_connection=,@@session.collation_server=/*!*/;
SET @@session.lc_time_names=/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
ALTER USER 'zlm'@'192.168.56.%' IDENTIFIED WITH 'mysql_native_password' AS '*512FB3FAA6F522E351929CAC70AD8EBB0F5901B6'
/*!*/;
# at
# :: server id end_log_pos CRC32 0x0882ea24 Rotate to mysql-bin. pos:
SET @@SESSION.GTID_NEXT= 'AUTOMATIC' /* added by mysqlbinlog */ /*!*/;
DELIMITER ;
# End of log file //I've known that the operation of altering password is in the uuid:12715627.Therefore,I could set gtid_purged to the previous gtid(uuid:5-12715626).
I tried to set "gtid_purged" to the right value according to the content of binlog on master.
root@localhost:mysql.sock [(none)]>set @@global.gtid_purged='2a4b3562-2ab6-11e8-be7a-080027de0e0e:5-12715626';
Query OK, rows affected (0.00 sec) root@localhost:mysql.sock [(none)]>start slave io_thread;
Query OK, rows affected (0.01 sec) root@localhost:mysql.sock [(none)]>show slave status\G
*************************** . row ***************************
Slave_IO_State:
Master_Host: 192.168.56.100
Master_User: repl
Master_Port:
Connect_Retry:
Master_Log_File: mysql-bin.
Read_Master_Log_Pos:
Relay_Log_File: relay-bin.
Relay_Log_Pos:
Relay_Master_Log_File: mysql-bin.
Slave_IO_Running: No //It was not running again.
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno:
Last_Error:
Skip_Counter:
Exec_Master_Log_Pos:
Relay_Log_Space:
Until_Condition: None
Until_Log_File:
Until_Log_Pos:
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno:
Last_IO_Error: Got fatal error from master when reading data from binary log: 'The slave is connecting using CHANGE MASTER TO MASTER_AUTO_POSITION = 1, but the master has purged binary logs containing GTIDs that the slave requires.'
Last_SQL_Errno:
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id:
Master_UUID: 2a4b3562-2ab6-11e8-be7a-080027de0e0e
Master_Info_File: /data/mysql/mysql3306/data/master.info
SQL_Delay:
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count:
Master_Bind:
Last_IO_Error_Timestamp: ::
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set:
Executed_Gtid_Set: 2a4b3562-2ab6-11e8-be7a-080027de0e0e:-
Auto_Position:
Replicate_Rewrite_DB:
Channel_Name:
Master_TLS_Version:
row in set (0.00 sec) root@localhost:mysql.sock [(none)]>reset master;
Query OK, rows affected (0.03 sec) root@localhost:mysql.sock [(none)]>set @@global.gtid_purged='2a4b3562-2ab6-11e8-be7a-080027de0e0e:1-12715626'; //I began with the first gtid this time.
Query OK, rows affected (0.00 sec) root@localhost:mysql.sock [(none)]>stop slave;
Query OK, rows affected (0.00 sec) root@localhost:mysql.sock [(none)]>start slave;
Query OK, rows affected (0.00 sec) root@localhost:mysql.sock [(none)]>show slave status\G
*************************** . row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.56.100
Master_User: repl
Master_Port:
Connect_Retry:
Master_Log_File: mysql-bin.
Read_Master_Log_Pos:
Relay_Log_File: relay-bin.
Relay_Log_Pos:
Relay_Master_Log_File: mysql-bin.
Slave_IO_Running: Yes //IO_Thread could work normally.
Slave_SQL_Running: No //SQL_Thread was down instead.
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: //A new arrival error 1396 occurred.
Last_Error: Error 'Operation ALTER USER failed for 'zlm'@'192.168..%'' on query. Default database: ''. Query: 'ALTER USER 'zlm'@'192.168..%' IDENTIFIED WITH 'mysql_native_password' AS '*512FB3FAA6F522E351929CAC70AD8EBB0F5901B6''
Skip_Counter:
Exec_Master_Log_Pos:
Relay_Log_Space:
Until_Condition: None
Until_Log_File:
Until_Log_Pos:
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno:
Last_IO_Error:
Last_SQL_Errno:
Last_SQL_Error: Error 'Operation ALTER USER failed for 'zlm'@'192.168..%'' on query. Default database: ''. Query: 'ALTER USER 'zlm'@'192.168..%' IDENTIFIED WITH 'mysql_native_password' AS '*512FB3FAA6F522E351929CAC70AD8EBB0F5901B6''
Replicate_Ignore_Server_Ids:
Master_Server_Id:
Master_UUID: 2a4b3562-2ab6-11e8-be7a-080027de0e0e
Master_Info_File: /data/mysql/mysql3306/data/master.info
SQL_Delay:
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State:
Master_Retry_Count:
Master_Bind:
Last_IO_Error_Timestamp:
Last_SQL_Error_Timestamp: ::
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set: 2a4b3562-2ab6-11e8-be7a-080027de0e0e:
Executed_Gtid_Set: 2a4b3562-2ab6-11e8-be7a-080027de0e0e:-
Auto_Position:
Replicate_Rewrite_DB:
Channel_Name:
Master_TLS_Version:
row in set (0.00 sec) root@localhost:mysql.sock [(none)]>ALTER USER 'zlm'@'192.168.56.%' IDENTIFIED WITH 'mysql_native_password' AS '*512FB3FAA6F522E351929CAC70AD8EBB0F5901B6'; //I couldn't executed it in mysql cli either.
ERROR (HY000): Operation ALTER USER failed for 'zlm'@'192.168.56.%'
root@localhost:mysql.sock [(none)]>
Then,I checked the user on both master and slave.I found that there was no "zlm" user on slave at all.
//Master
root@localhost:mysql3306.sock [(none)]>select user,host from mysql.user;
+---------------+--------------+
| user | host |
+---------------+--------------+
| repl | 192.168..% |
| zlm | 192.168..% |
| bkuser | localhost |
| mysql.session | localhost |
| mysql.sys | localhost |
| root | localhost |
+---------------+--------------+
rows in set (0.00 sec) //Slave
root@localhost:mysql.sock [(none)]>select user,host from mysql.user;
+---------------+--------------+
| user | host |
+---------------+--------------+
| repl | 192.168..% |
| bkuser | localhost |
| mysql.session | localhost |
| mysql.sys | localhost |
| root | localhost |
+---------------+--------------+
rows in set (0.00 sec) root@localhost:mysql.sock [(none)]>create user zlm@'192.168.56.%'; //I created the absent user "zlm" on slave.
Query OK, rows affected (0.00 sec) root@localhost:mysql.sock [(none)]>start slave sql_thread;
Query OK, rows affected (0.01 sec) root@localhost:mysql.sock [(none)]>show slave status\G
*************************** . row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.56.100
Master_User: repl
Master_Port:
Connect_Retry:
Master_Log_File: mysql-bin.
Read_Master_Log_Pos:
Relay_Log_File: relay-bin.
Relay_Log_Pos:
Relay_Master_Log_File: mysql-bin.
Slave_IO_Running: Yes
Slave_SQL_Running: Yes //It was ok.
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno:
Last_Error:
Skip_Counter:
Exec_Master_Log_Pos:
Relay_Log_Space:
Until_Condition: None
Until_Log_File:
Until_Log_Pos:
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master:
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno:
Last_IO_Error:
Last_SQL_Errno:
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id:
Master_UUID: 2a4b3562-2ab6-11e8-be7a-080027de0e0e
Master_Info_File: /data/mysql/mysql3306/data/master.info
SQL_Delay:
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
Master_Retry_Count:
Master_Bind:
Last_IO_Error_Timestamp:
Last_SQL_Error_Timestamp:
Master_SSL_Crl:
Master_SSL_Crlpath:
Retrieved_Gtid_Set: 2a4b3562-2ab6-11e8-be7a-080027de0e0e:
Executed_Gtid_Set: 27af30ca--11e8-ad7e-080027de0e0e:, //This gtid was generated by the statment of "create user ... ".
2a4b3562-2ab6-11e8-be7a-080027de0e0e:- //The executed gtid was equal with the master.
Auto_Position:
Replicate_Rewrite_DB:
Channel_Name:
Master_TLS_Version:
row in set (0.00 sec)
Summary
- Error 1236 is common phenomenon in our replication of MySQL which always due to lack of specific binlog files.
- Before you execute a statement on master.You'd better start your slave replication at first.Obviously it means our test environment.
- If you were not sure which transaction was stuck.Analyze the binlog file is always a good idea to solve these kind of issues.
- In my case,I know clearly about what I've operated in master.So I can easily execute the same operation on slave.But the product environment is complecated and it's not recommended to do so.
GTID环境中手动修复主从故障一例（Error 1236/Error 1396）的更多相关文章
- GTID环境中手动修复主从故障一例（Error 1146）
Preface In my last test of pt-heartbeat,both of master and slave were out of disk.And the my ...
- SqlServer 禁止架构更改的复制中手动修复使发布和订阅中分别增加的字段同步
原文:SqlServer 禁止架构更改的复制中手动修复使发布和订阅中分别增加的字段同步 由于之前的需要,禁止了复制架构更改,以至在发布中添加一个字段,并不会同步到订阅中,而现在又在订阅中添加了一个同名 ...
- 企业运维 | MySQL关系型数据库在Docker与Kubernetes容器环境中快速搭建部署主从实践
[点击 关注「 WeiyiGeek」公众号 ] 设为「️ 星标」每天带你玩转网络安全运维.应用开发.物联网IOT学习! 希望各位看友[关注.点赞.评论.收藏.投币],助力每一个梦想. 本章目录 目录 ...
- 故障案例 | 主从复制环境中tokudb引擎报错排查过程
欢迎来到 GreatSQL社区分享的MySQL技术文章,如有疑问或想学习的内容,可以在下方评论区留言,看到后会进行解答 GreatSQL社区原创内容未经授权不得随意使用,转载请联系小编并注明来源. 0 ...
- 业务零影响！如何在Online环境中巧用MySQL传统复制技术【转】
业务零影响!如何在Online环境中巧用MySQL传统复制技术 这篇文章我并不会介绍如何部署一个MySQL复制环境或keepalived+双主环境,因为此类安装搭建的文章已经很多,大家也很熟悉.在这篇 ...
- .NET 环境中使用RabbitMQ RabbitMQ与Redis队列对比 RabbitMQ入门与使用篇
.NET 环境中使用RabbitMQ 在企业应用系统领域,会面对不同系统之间的通信.集成与整合,尤其当面临异构系统时,这种分布式的调用与通信变得越发重要.其次,系统中一般会有很多对实时性要求不高的 ...
- 5.7 并行复制配置 基于GTID 搭建中从 基于GTID的备份与恢复，同步中断处理
5.7 并行复制配置 基于GTID 搭建中从 基于GTID的备份与恢复,同步中断处理 这个文章包含三个部分 1:gtid的多线程复制2:同步中断处理3:GTID的备份与恢复 下面文字相关的东西 大部分 ...
- 生产环境中的kubernetes 优先级与抢占
kubernetes 中的抢占功能是调度器比较重要的feature,但是真正使用起来还是比较危险,否则很容易把低优先级的pod给无辜kill.为了提高GPU集群的资源利用率,决定勇于尝试一番该feat ...
- Redis 哨兵模式实现主从故障互切换
200 ? "200px" : this.width)!important;} --> 介绍 Redis Sentinel 是一个分布式系统, 你可以在一个架构中运行多个 S ...
随机推荐
- 如何将Twitter的内容导入到SAP CRM和C4C
Twitter的内容导入SAP CRM Interaction Center呼叫中心 具体步骤查看我的博客Twitter(also Facebook) is official integrated i ...
- 如何查看某个用户指定时间段的ABAP开发记录
输入用户名和想查询的时间段: 执行得到结果.双击可查看具体代码: 工具源代码: REPORT tool_dev_history. PARAMETERS: name TYPE usr02-bname O ...
- Ruby SSL_connect returned=1 errno=0 state=SSLv3 read server certificate B:
最近使用ruby-china的源连接不上 使用gem update遇到这个问题, 原来是ruby没有包含SSL证书,所以Https的链接被服务器拒绝. 解决方法很简单,首先在这里下载证书(http:/ ...
- Netbackup-Oracle数据库恢复演练(上)
目录 第一章NBU实施内容 1 第二章 环境准备 1 2.1添加LICENSE: 1 2.2安装配置NBU客户端(linux,unix相同) 2 2.3配置并验证oracle备份脚本 3 2.4添加备 ...
- Maven项目导出可执行jar
配置文件中添加插件 <plugin> <groupId>org.apache.maven.plugins</groupId> <artifactId>m ...
- CPP-STL：用vector保存对象时保存指针的优点, 以及reserve的使用（转）
代码1 #include <vector> #include <stdio.h> class A { public: A() { printf("A()/n" ...
- Java-笔记1
/* 对第一个java程序进行总结 1. java程序编写-编译-运行的过程 编写:我们将编写的java代码保存在以".java"结尾的源文件中 编译:使用javac.exe命令编 ...
- flexslider 图片轮播插件参数
$(window).load(function() { $('.flexslider').flexslider({ animation: "fade", //String: Sel ...
- 最近的阅读list
fast rcnn 统一了sppnet和rcnn,将原来rcnn分stage的训练合为一个整体的stage,一次完成cls, regression的训练.引入两个loss函数,一个是用来进行cls的, ...
- rcnn spp_net hcp
rcnn开创性工作,但是计算时间太长,重复计算太大. spp_net将重复计算避免了. hcp是yan shuicheng那边的,是用bing生成regions,然后用normalized cut将这 ...