alice-yang94 opened a new issue #14591: URL: https://github.com/apache/shardingsphere/issues/14591
## Bug Report ### Which version of ShardingSphere did you use? - ShardingSphere 5.1.0 ### Which project did you use? ShardingSphere-JDBC or ShardingSphere-Proxy? - ShardingSphere-Proxy ### Background environment When using MogDB_v2.1.0.3 (same as openGauss 2.1.0) to set up one primary database, two standby databases and using switchover or failover on one of the standby database. After switchover or failover, the current standby database is switched to primary as the new primary database. The original primary database is switched as a standby database. ### Expected behavior ShardingSphere proxy's discovery module should recognise the new primary database and continue with new primary and standby databases. ### Actual behavior ShardingSphere proxy's discovery module does not recognise the change of new primary database - For switchover, ShardingSphere proxy has error: An exception occurred while find primary data source url. When trying to connect and execute sql DDL commands, ShardingSphere proxy has error：cannot execute UPDATE in a read-only transaction. - For failover, ShardingSphere proxy has error: Connection is not available. The error log keeps trying to connect to the old primary database and cause the connection fail ### Reason analyze (If you can) ShardingSphere proxy's discovery module does not recognise the change of new primary database. ### Steps to reproduce the behavior, such as: SQL to execute, sharding rule configuration, when exception occur etc. 1. environment: - MogDB_v2.1.0.3 (release_MogDB_openEuler_ARM64 和 release_MogDB_CentOS_x86） - Zookeeper 3.7.0 - ShardingSphere 5.1.0 - sysbench: https://github.com/akopytov/sysbench/tree/master/src 2. set up 1 primary, 2 standby databases according to mogdb docs 3. start zookeeper and sharding sphere proxy ``` [wenke@node156 wenke]$ cat apache-shardingsphere-5.1.0-shardingsphere-proxy-bin/conf/config-db-discovery.yaml schemaName: rwsplit_test_db dataSources: replica_ds_0: url: jdbc:postgresql://172.16.3.156:12000/postgres?serverTimezone=UTC&useSSL=false username: test1 password: Qwer1234 connectionTimeoutMilliseconds: 300 idleTimeoutMilliseconds: 600 maxLifetimeMilliseconds: 1200 maxPoolSize: 1000 minPoolSize: 200 replica_ds_1: url: jdbc:postgresql://172.16.3.156:13000/postgres?serverTimezone=UTC&useSSL=false username: test1 password: Qwer1234 connectionTimeoutMilliseconds: 300 idleTimeoutMilliseconds: 600 maxLifetimeMilliseconds: 1200 maxPoolSize: 1000 minPoolSize: 200 replica_ds_2: url: jdbc:postgresql://172.16.3.156:14000/postgres?serverTimezone=UTC&useSSL=false username: test1 password: Qwer1234 connectionTimeoutMilliseconds: 300 idleTimeoutMilliseconds: 600 maxLifetimeMilliseconds: 1200 maxPoolSize: 1000 minPoolSize: 200 rules: - !READWRITE_SPLITTING dataSources: rd_rs: autoAwareDataSourceName: pr_ds - !DB_DISCOVERY dataSources: pr_ds: dataSourceNames: - replica_ds_0 - replica_ds_1 - replica_ds_2 discoveryHeartbeatName: og-heartbeat discoveryTypeName: og discoveryHeartbeats: og-heartbeat: props: keep-alive-cron: '0/5 * * * * ?' discoveryTypes: og: type: openGauss props: group-name: 92504d5b-6dec-11e8-91ea-246e9612aaf1 ``` 4. sysbench command executed: ``` sysbench oltp_read_write --db-driver=pgsql --pgsql-host=localhost --pgsql-port=3307 --pgsql-user=test1 --pgsql-password=Qwer1234 --pgsql-db=rwsplit_test_db --skip_trx --tables=10 --table-size=1000000 --threads=64 --report-interval=1 --time=1800 run ``` 5. execute switchover or failover 5.1 switchover ``` [standby1]$ bin/gs_ctl -D data_dir switchover ``` ss switchover error screenshots:   5.2 failover ``` [primary]$ bin/gs_ctl -D data_dir stop [standby1]$ bin/gs_ctl -D data_dir failover ``` ss failover error screenshots:  -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
