溫馨提示×
您好，登錄后才能下訂單哦！
點擊 登錄注冊 即表示同意《億速云用戶服務條款》
您好，登錄后才能下訂單哦！
Oracle 12C Sharding 部署和測試
日期： 2019-02-26
作者：陳舉超
目錄
一：環境說明 ... 4
二：安裝GSM/Shard catalog/ORACLE 12.2 RDBMS . 5
2.1 Gsm 安裝 ... 5
2.2 ORACLE 12.2 RDBMS 安裝 ... 8
2.3 創建并啟動監聽 ... 8
2.4 創建Shard Catalog database . 8
三：配置GSM/Shard director . 8
3.1 解鎖用戶 ... 8
3.2 創建管理用戶mygds . 8
3.3 啟動listener . 8
3.4 創建shard catalog 配置remote scheduler agent . 8
3.5 創建和啟動shard director. 9
3.6 添加操作系統認證. 9
3.7 在所有的shard 節點分別執行Agent . 10
3.8 創建shard 數據庫的數據文件存儲路徑 ... 11
四：創建System-Managed SDB . 11
4.1 連接gsm .. 11
4.2 添加shardgroup . 11
4. 3 創建shard . 11
4.4 檢查配置 ... 12
4.5 部署/deploy . 13
4.6 檢查配置信息 ... 14
4.7 創建service . 16
五：創建用戶和對象 ... 16
5.1 在catalog 數據庫中創建業務用戶 ... 16
5.2 創建表空間集合 ... 16
5.3 為duplicated tables 創建表空間 ... 17
5.4 創建 root 表 Customers . 17
5.5 創建其他 sharded table Orders. 18
5.6 為 OrderId 列創建序列 ... 18
5.7 創建 SHARDED TABLE LineItems . 18
5.8 創建 duplicated tables. 18
5.9 檢查是否有錯誤 ... 19
六：驗證環境- 表空間/chunks . 20
6.1 在db01 檢查chunks 信息 ... 20
6.2 在 sd1,sd2 節點檢查表空間和 chunks 信息 ... 20
6.3 在catalog 數據庫檢查chunks 信息 ... 23
6.4 驗證環境 ... 23
七：訪問Shard . 24
7.1 訪問單獨一個shard . 24
7.2 訪問多個shard . 26
八：shard 數據庫啟動和停止 ... 27
8.1 關閉 ... 27
8.2 啟動
九：常見問題 ... 29
9.1 create shard –shardgroup 錯誤 ... 29
9.2 catalog 創建用戶和表，不能同步到 shard db 庫 ... 29
9.3 GSM 啟動失敗，顯示異常
十：參考文獻
一：環境說明
OS ：CentOS Linux release 7.5
DB ：Oracle 12.2.0.1.0
GSMOCI ：2.2.1
本次使用三臺虛擬機進行安裝：
其中：
(1) db01 主機上安裝Shard Director 和Shard catalog ，也就是分別安裝GSM 、ORACLE 12.2 RDBMS 、non-cdb Database(GSM 和ORACLE 12.2 環境變量需要手動切換) ；
(2)db02 主機上安裝Shard1 ，也就是ORACLE 12.2 RDBMS ，不需要創建數據庫；
(3) db03 主機上安裝Shard2 ，也就是ORACLE 12.2 RDBMS ，不需要創建數據庫；
|
主機 |
組件 |
SID |
Oracle_Home |
|
db01 |
Shard Director |
|
/u01/app/oracle/product/12.2.0/gsm_1 |
|
db01 |
Shard catalog |
db01 |
/u01/app/oracle/product/12.2.0/db_1 |
|
db02 |
Shard1 |
sh2 |
/u01/app/oracle/product/12.2.0/db_1 |
|
db03 |
Shard2 |
sh3 |
/u01/app/oracle/product/12.2.0/db_1 |
安裝文件下載鏈接如下：
https://www.oracle.com/technetwork/cn/database/options/partitioning/downloads/index.html
Oracle 軟件下載
GSM 下載
略
Netca
在Shard catalog 服務器gsm1 創建 non-cdb 數據庫。創建過程與普通數據庫相同。
三：配置GSM/Shard director
https://blogs.oracle.com/database4cn/12c-oracle-sharding
在db01 服務器(catalog 數據庫/shard director) ，連接到Sharding catalog 數據庫, 解鎖 GSMCATUSER 用戶，shard director 通過GSMCATUSER 用戶連接到shard catalog database
SQL> alter user gsmcatuser identified by oracle account unlock;
在 catalog 數據庫，創建管理用戶mygds ，用戶mygds 用于存儲Sharding 管理信息，GDSCTL 接口通過用戶mygds 連接到catalog 數據庫。
SQL> create user mygds identified by oracle;
SQL> grant connect, create session, gsmadmin_role to mygds;
SQL> grant inherit privileges on user SYS to GSMADMIN_INTERNAL;
在db01 服務器(catalog 數據庫/shard director) ，啟動listener
在db01 服務器(catalog 數據庫/shard director) ，創建shard catalog ，在shard catalog 中配置remote scheduler agent.
參數含義：
-user : 指定管理用戶，在前面步驟中創建的catalog database 管理用戶mygds
-database : 指定catalog database 信息，catalog 數據庫的主機名: 監聽器port: catalog 數據庫db_name
-sdb : 指定sharded database name
-agent_port: 設置端口，用于shard 節點agent 連接到GSM
-agent_password: 設置密碼，用于shard 節點agent 連接到GSM
如果沒有指定- sharding 參數，默認是創建system-managed (default) 類型
[oracle@db01 ~]$ export ORACLE_BASE=/u01/app/oracle
[oracle@db01 ~]$ export ORACLE_HOME=/u01/app/oracle/products/12.2.0/gsm_1
[oracle@db01 ~]$
export PATH=/u01/app/oracle/products/12.2.0/gsmh_11/bin:$PATH:$HOME/bin
[oracle@db01 ~]$ gdsctl
GDSCTL: Version 12.2.0.1.0 - Production on Mon Feb 25 10:38:54 CST 2019
Copyright (c) 2011, 2016, Oracle. All rights reserved.
Welcome to GDSCTL, type "help" for information.
Warning: current GSM name is not set automatically because gsm.ora contains zero or several GSM entries. Use "set gsm" command to set GSM for the session.
Current GSM is set to GSMORA
GDSCTL> create shardcatalog -database db01:1521:db01 -chunks 12 -user mygds/oracle -sdb shdb -region region1, region2 -agent_port 8080 -agent_password oracle
Catalog is created
參數含義：
-gsm: 指定shard director 名稱
-listener: 指定shard director 的監聽端口，注意不能與數據庫的listener 端口沖突
-catalog: 指定catalog database 信息，catalog 數據庫的主機名: 監聽器port: catalog 數據庫db_name
GDSCTL> add gsm -gsm sharddirector1 -listener 1522 -pwd oracle -catalog db01:1521:db01 -region region1
GSM successfully added
GDSCTL> start gsm -gsm sharddirector1
GSM 已成功啟動
GDSCTL> add credential -credential cre_reg1 -osaccount oracle -ospassword oracle
操作已成功完成
--- 在db02 節點執行
[oracle@db02 ~]$ schagent -start
Scheduler agent started using port 24409
[oracle@db02 ~]$ schagent -status
Agent running with PID 8537
Agent_version:12.2.0.1.2
Running_time:00:00:11
Total_jobs_run:0
Running_jobs:0
Platform:Linux
ORACLE_HOME:/u01/app/oracle/product/12.2.0/db_1
ORACLE_BASE:/u01/app/oracle
Port:24409
Host:db02
--- 在db03 節點執行
[oracle@db03 ~]$ schagent -start
Scheduler agent started using port 19249
[oracle@db03 ~]$ schagent -status
Agent running with PID 8344
Agent_version:12.2.0.1.2
Running_time:00:00:05
Total_jobs_run:0
Running_jobs:0
Platform:Linux
ORACLE_HOME:/u01/app/oracle/product/12.2.0/db_1
ORACLE_BASE:/u01/app/oracle
Port:19249
Host:db03
----- 密碼oracle 和端口8080
[oracle@db02 ~]$ echo oracle | schagent -registerdatabase 192.168.2.226 8080
Agent Registration Password ?
Oracle Scheduler Agent Registration for 12.2.0.1.2 Agent
Agent Registration Successful!
[oracle@db03 ~]$ echo oracle | schagent -registerdatabase 192.168.2.226 8080
Agent Registration Password ?
Oracle Scheduler Agent Registration for 12.2.0.1.2 Agent
Agent Registration Successful!
[oracle@db02 ~]$ mkdir /u01/app/oracle/oradata
[oracle@db02 ~]$ mkdir /u01/app/oracle/fast_recovery_area
[oracle@db03 ~]$ mkdir /u01/app/oracle/oradata
[oracle@db03 ~]$ mkdir /u01/app/oracle/fast_recovery_area
在Shard 服務器 sd1 連接到shard director/GSM 服務器(db01)
[oracle@db02 ~]$ ssh oracle@db01
……
oracle@db01's password:
Last login: Mon Feb 25 10:45:20 2019 from 192.168.70.249
[oracle@db01 ~]$ source .bash_profile
[oracle@db01 ~]$ gdsctl
GDSCTL: 版本 12.2.0.1.0 - Production, 時間: Mon Feb 25 11:16:04 CST 2019
版權所有 (c) 2011, 2016, Oracle 。保留所有權利。
歡迎使用GDSCTL, 請鍵入 "help" 以獲得信息。
當前 GSM 設置為 SHARDDIRECTOR1
GDSCTL> set gsm -gsm sharddirector1
GDSCTL>connect mygds/oracle
已建立目錄連接
shardgroup 是一組shard 的集合，shardgroup 名稱為primary_shardgroup ，-deploy_as primary 表示這個group 中的shard 都是主庫。
GDSCTL> add shardgroup -shardgroup primary_shardgroup -deploy_as primary -region region1
操作已成功完成
將每個shard 地址添加到catalog 的valid node checking for registration (VNCR) 列表，并且創建shard
GDSCTL> add invitednode db02
GDSCTL> create shard -shardgroup primary_shardgroup -destination db02 -credential cre_reg1 -sys_password oracle
操作已成功完成
數據庫的唯一名稱: sh2
GDSCTL> add invitednode db03
GDSCTL> create shard -shardgroup primary_shardgroup -destination db03 -credential cre_reg1 -sys_password oracle
操作已成功完成
數據庫的唯一名稱: sh3
GDSCTL>config
區域
------------------------
region1
region2
GSM
------------------------
sharddirector1
sharddirector3
分片數據庫
------------------------
shdb
數據庫
------------------------
sh2
sh3
分片組
------------------------
primary_shardgroup
分片空間
------------------------
shardspaceora
服務
------------------------
GDSCTL 暫掛請求
------------------------
命令 對象 狀態
-- -- --
全局屬性
------------------------
名稱: oradbcloud
主 GSM: sharddirector1
DDL 序列號: 0
GDSCTL> config shardspace
分片空間 塊
---- -
shardspaceora 12
GDSCTL> config shardgroup
分片組 塊 區域 分片空間
--- - -- ----
primary_shardgroup 12 region1 shardspaceora
GDSCTL> config vncr
名稱 組 ID
-- ----
192.168.2.226
db02
db03
GDSCTL> config shard
名稱 分片組 狀態 狀態 區域 可用性
-- --- -- -- -- ---
sh2 primary_shardgroup U 無 region1 -
sh3 primary_shardgroup U 無 region1 -
執行deploy 后，自動在db02 和db03 服務器上采用靜默安裝方式部署shard 數據庫，速度較慢；
GDSCTL> deploy
deploy: examining configuration...
deploy: deploying primary shard 'sh2' ...
deploy: network listener configuration successful at destination 'db02'
deploy: starting DBCA at destination 'db02' to create primary shard 'sh2' ...
deploy: deploying primary shard 'sh3' ...
deploy: network listener configuration successful at destination 'db03'
deploy: starting DBCA at destination 'db03' to create primary shard 'sh3' ...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: waiting for 2 DBCA primary creation job(s) to complete...
deploy: DBCA primary creation job succeeded at destination 'db02' for shard 'sh2'
deploy: waiting for 1 DBCA primary creation job(s) to complete...
deploy: waiting for 1 DBCA primary creation job(s) to complete...
deploy: DBCA primary creation job succeeded at destination 'db03' for shard 'sh3'
deploy: requesting Data Guard configuration on shards via GSM
deploy: shards configured successfully
操作已成功完成
GDSCTL> config shard
名稱 分片組 狀態 狀態 區域 可用性
-- --- -- -- -- ---
sh2 primary_shardgroup 確定 已部署 region1 ONLINE
sh3 primary_shardgroup 確定 已部署 region1 ONLINE
GDSCTL> databases
數據庫: "sh2" 注冊時間: Y 狀態: 確定 ONS: N 。角色: PRIMARY 實例: 1 區域: region1
已注冊的實例:
shdb%1
數據庫: "sh3" 注冊時間: Y 狀態: 確定 ONS: N 。角色: PRIMARY 實例: 1 區域: region1
已注冊的實例:
shdb%11
GDSCTL> config shard -shard sh2
名稱: sh2
分片組: primary_shardgroup
狀態: 確定
狀態: 已部署
區域: region1
連接字符串: db02:1521/sh2:dedicated
SCAN 地址:
ONS 遠程端口: 0
磁盤閾值, 毫秒: 20
CPU 閾值, %: 75
版本: 12.2.0.0
上次失敗的 DDL:
DDL 錯誤: ---
失敗的 DDL ID:
可用性: ONLINE
Rack:
支持的服務
------------------------
名稱 首選 狀態
-- -- --
GDSCTL> config shard -shard sh3
名稱: sh3
分片組: primary_shardgroup
狀態: 確定
狀態: 已部署
區域: region1
連接字符串: db03:1521/sh3:dedicated
SCAN 地址:
ONS 遠程端口: 0
磁盤閾值, 毫秒: 20
CPU 閾值, %: 75
版本: 12.2.0.0
上次失敗的 DDL:
DDL 錯誤: ---
失敗的 DDL ID:
可用性: ONLINE
Rack:
支持的服務
------------------------
名稱 首選 狀態
-- -- --
GDSCTL> add service -service oltp_rw_srvc -role primary
操作已成功完成
GDSCTL> start service -service oltp_rw_srvc
操作已成功完成
GDSCTL> status service
服務 "oltp_rw_srvc.shdb.oradbcloud" 具有 2 個實例。關系: ANYWHERE
實例 "shdb%1", 名稱: "sh2", 數據庫: "sh2", 區域: "region1", 狀態: ready 。
實例 "shdb%11", 名稱: "sh3", 數據庫: "sh3", 區域: "region1", 狀態: ready 。
---db01
SQL> alter session enable shard ddl;
SQL> create user app_schema identified by oracle;
SQL> grant all privileges to app_schema;
SQL> grant gsmadmin_role to app_schema;
SQL> grant select_catalog_role to app_schema;
SQL> grant connect, resource to app_schema;
SQL> grant dba to app_schema;
SQL> grant execute on dbms_crypto to app_schema;
---db01
SQL> conn app_schema
Enter password:
Connected.
SQL> alter session enable shard ddl;
Session altered.
SQL> create tablespace set tsp_set_1 using template (datafile size 100m autoextend on next 10m maxsize unlimited extent management local segment space management auto);
Tablespace created.
GDSCTL> set gsm -gsm sharddirector1
GDSCTL> config shard
已建立目錄連接
名稱 分片組 狀態 狀態 區域 可用性
-- --- -- -- -- ---
sh2 primary_shardgroup 確定 DDL 錯誤 region1 ONLINE
sh3 primary_shardgroup 確定 DDL 錯誤 region1 ONLINE
--- 這個測試中duplicated table 是Products table.
SQL> CREATE TABLESPACE products_tsp datafile '/u01/app/oracle/oradata/db01/products_tsp01.dbf' size 100m autoextend on next 10M maxsize unlimited extent management local uniform size 1m;
Tablespace created.
SQL > CONNECT app_schema / oracle
SQL > ALTER SESSION ENABLE SHARD DDL ;
SQL > CREATE SHARDED TABLE Customers
(
CustId VARCHAR2 ( 60 ) NOT NULL ,
FirstName VARCHAR2 ( 60 ),
LastName VARCHAR2 ( 60 ),
Class VARCHAR2 ( 10 ),
Geo VARCHAR2 ( 8 ),
CustProfile VARCHAR2 ( 4000 ),
Passwd RAW ( 60 ),
CONSTRAINT pk_customers PRIMARY KEY ( CustId ),
CONSTRAINT json_customers CHECK ( CustProfile IS JSON )
) TABLESPACE SET TSP_SET_1
PARTITION BY CONSISTENT HASH ( CustId ) PARTITIONS AUTO ;
SQL > CREATE SHARDED TABLE Orders
(
OrderId INTEGER NOT NULL ,
CustId VARCHAR2 ( 60 ) NOT NULL ,
OrderDate TIMESTAMP NOT NULL ,
SumTotal NUMBER ( 19 , 4 ),
Status CHAR ( 4 ),
CONSTRAINT pk_orders PRIMARY KEY ( CustId , OrderId ),
CONSTRAINT fk_orders_parent FOREIGN KEY ( CustId )
REFERENCES Customers ON DELETE CASCADE
) PARTITION BY REFERENCE ( fk_orders_parent );
SQL > CREATE SEQUENCE Orders_Seq ;
Sequence created.
SQL > CREATE SHARDED TABLE LineItems
(
OrderId INTEGER NOT NULL ,
CustId VARCHAR2 ( 60 ) NOT NULL ,
ProductId INTEGER NOT NULL ,
Price NUMBER ( 19 , 4 ),
Qty NUMBER ,
CONSTRAINT pk_items PRIMARY KEY ( CustId , OrderId , ProductId ),
CONSTRAINT fk_items_parent FOREIGN KEY ( CustId , OrderId )
REFERENCES Orders ON DELETE CASCADE
) PARTITION BY REFERENCE ( fk_items_parent );
---In this example , the Products table is a duplicated object.
SQL > CREATE DUPLICATED TABLE Products
(
ProductId INTEGER GENERATED BY DEFAULT AS IDENTITY PRIMARY KEY ,
Name VARCHAR2 ( 128 ),
DescrUri VARCHAR2 ( 128 ),
LastPrice NUMBER ( 19 , 4 )
) TABLESPACE products_tsp ;
GDSCTL> set gsm -gsm sharddirector1
GDSCTL> show ddl
ID DDL 文本 失敗的分片
-- ------ -----
108 grant dba to app_schema
109 grant execute on dbms_crypto to app_s...
110 CREATE TABLESPACE SET TSP_SET_1 using...
111 CREATE TABLESPACE products_tsp dataf...
112 CREATE SHARDED TABLE Customers ( ...
113 CREATE SHARDED TABLE Orders ( O...
114 CREATE SHARDED TABLE LineItems ( ...
115 CREATE MATERIALIZED VIEW "APP_SCHEMA"...
116 CREATE OR REPLACE FUNCTION PasswCreat...
117 CREATE OR REPLACE FUNCTION PasswCheck...
GDSCTL> config shard -shard sh2
名稱: sh2
分片組: primary_shardgroup
狀態: 確定
狀態: 已部署
區域: region1
連接字符串: db02:1521/sh2:dedicated
SCAN 地址:
ONS 遠程端口: 0
磁盤閾值, 毫秒: 20
CPU 閾值, %: 75
版本: 12.2.0.0
上次失敗的 DDL: CREATE MATERIALIZED VIEW "APP_...
DDL 錯誤: ORA-00942: table or view does not exist
ORA-06512: at "SYS.EXECASUSER", line 44
ORA-06512: at "SYS.DBMS_SYS_SQL", line 1529
ORA-06512: at "SYS.EXECASUSER", line 31
ORA-06512: at "GSMADMIN_INTERNAL.EXECUTEDDL", line 70
ORA-06512: at line 1 \(ngsmoci_execute\)
失敗的 DDL ID: 57
可用性: ONLINE
Rack:
支持的服務
------------------------
名稱 首選 狀態
-- -- --
oltp_rw_srvc 是 啟用
前面創建shardcatalog時指定chunks為12，因此后續創建shard table分配12個chunks
GDSCTL>config chunks
塊
------------------------
數據庫 自 至
--- - -
sh2 1 6
sh3 7 12
---db02
--表空間
SQL> select TABLESPACE_NAME, BYTES/1024/1024 MB from sys.dba_data_files order by tablespace_name;
TABLESPACE_NAME MB
------------------------------ ----------
C001TSP_SET_1 100
C002TSP_SET_1 100
C003TSP_SET_1 100
C004TSP_SET_1 100
C005TSP_SET_1 100
C006TSP_SET_1 100
PRODUCTS_TSP 100
SYSAUX 480
SYSTEM 810
TSP_SET_1 100
UNDOTBS1 70
TABLESPACE_NAME MB
------------------------------ ----------
USERS 5
12 rows selected.
創建了6個表空間，分別是C001TSP_SET_1 ~ 表空間C006TSP_SET_1，因為設置chunks=12，每個shard有6個chunks。
每個表空間有一個datafile，大小是100M，這個是在創建tablespace set時設置的datafile 100M。
--檢查chunks
SQL> set linesize 140
SQL> column table_name format a20
SQL> column tablespace_name format a20
SQL> column partition_name format a20
SQL> show parameter db_unique_name
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
db_unique_name string sh2
SQL> select table_name, partition_name, tablespace_name from dba_tab_partitions
where tablespace_name like 'C%TSP_SET_1' order by tablespace_name; 2
TABLE_NAME PARTITION_NAME TABLESPACE_NAME
-------------------- -------------------- --------------------
LINEITEMS CUSTOMERS_P1 C001TSP_SET_1
CUSTOMERS CUSTOMERS_P1 C001TSP_SET_1
ORDERS CUSTOMERS_P1 C001TSP_SET_1
CUSTOMERS CUSTOMERS_P2 C002TSP_SET_1
ORDERS CUSTOMERS_P2 C002TSP_SET_1
LINEITEMS CUSTOMERS_P2 C002TSP_SET_1
CUSTOMERS CUSTOMERS_P3 C003TSP_SET_1
LINEITEMS CUSTOMERS_P3 C003TSP_SET_1
ORDERS CUSTOMERS_P3 C003TSP_SET_1
LINEITEMS CUSTOMERS_P4 C004TSP_SET_1
CUSTOMERS CUSTOMERS_P4 C004TSP_SET_1
ORDERS CUSTOMERS_P4 C004TSP_SET_1
CUSTOMERS CUSTOMERS_P5 C005TSP_SET_1
ORDERS CUSTOMERS_P5 C005TSP_SET_1
LINEITEMS CUSTOMERS_P5 C005TSP_SET_1
CUSTOMERS CUSTOMERS_P6 C006TSP_SET_1
ORDERS CUSTOMERS_P6 C006TSP_SET_1
LINEITEMS CUSTOMERS_P6 C006TSP_SET_1
18 rows selected.
在sd2節點檢查表空間和chunks信息
--表空間
SQL> select TABLESPACE_NAME, BYTES/1024/1024 MB from sys.dba_data_files order by tablespace_name;
TABLESPACE_NAME MB
------------------------------ ----------
C007TSP_SET_1 100
C008TSP_SET_1 100
C009TSP_SET_1 100
C00ATSP_SET_1 100
C00BTSP_SET_1 100
C00CTSP_SET_1 100
PRODUCTS_TSP 100
SYSAUX 480
SYSTEM 810
TSP_SET_1 100
UNDOTBS1 65
USERS 5
12 rows selected.
創建了6個表空間，分別是C007TSP_SET_1 ~ 表空間C00CTSP_SET_1，因為設置chunks=12，每個shard有6個chunks。
每個表空間有一個datafile，大小是100M，這個是在創建tablespace set時設置的datafile 100M。
--檢查chunks
SQL> set linesize 140
SQL> column table_name format a20
SQL> column tablespace_name format a20
SQL> column partition_name format a20
SQL> select table_name, partition_name, tablespace_name from dba_tab_partitions
where tablespace_name like 'C%TSP_SET_1' order by tablespace_name;
TABLE_NAME PARTITION_NAME TABLESPACE_NAME
-------------------- -------------------- --------------------
ORDERS CUSTOMERS_P7 C007TSP_SET_1
LINEITEMS CUSTOMERS_P7 C007TSP_SET_1
CUSTOMERS CUSTOMERS_P7 C007TSP_SET_1
ORDERS CUSTOMERS_P8 C008TSP_SET_1
CUSTOMERS CUSTOMERS_P8 C008TSP_SET_1
LINEITEMS CUSTOMERS_P8 C008TSP_SET_1
LINEITEMS CUSTOMERS_P9 C009TSP_SET_1
ORDERS CUSTOMERS_P9 C009TSP_SET_1
CUSTOMERS CUSTOMERS_P9 C009TSP_SET_1
LINEITEMS CUSTOMERS_P10 C00ATSP_SET_1
ORDERS CUSTOMERS_P10 C00ATSP_SET_1
CUSTOMERS CUSTOMERS_P10 C00ATSP_SET_1
ORDERS CUSTOMERS_P11 C00BTSP_SET_1
LINEITEMS CUSTOMERS_P11 C00BTSP_SET_1
CUSTOMERS CUSTOMERS_P11 C00BTSP_SET_1
LINEITEMS CUSTOMERS_P12 C00CTSP_SET_1
CUSTOMERS CUSTOMERS_P12 C00CTSP_SET_1
ORDERS CUSTOMERS_P12 C00CTSP_SET_1
18 rows selected.
SQL> set echo off
SQL> select a.name Shard, count( b.chunk_number) Number_of_Chunks from
gsmadmin_internal.database a, gsmadmin_internal.chunk_loc b where
a.database_num=b.database_num group by a.name;
SHARD NUMBER_OF_CHUNKS
------------------------------ ----------------
sh2 6
sh3 6
--catalog數據庫
SQL> conn app_schema/oracle
Connected.
SQL> select table_name from user_tables;
TABLE_NAME
--------------------------------------------------------------------------------
PRODUCTS
MLOG$_PRODUCTS
CUSTOMERS
ORDERS
LINEITEMS
RUPD$_PRODUCTS
6 rows selected.
--shard節點sd1和sd2
SQL> conn app_schema/oracle
Connected.
SQL> select table_name from user_tables;
TABLE_NAME
--------------------
CUSTOMERS
ORDERS
LINEITEMS
PRODUCTS
在連接串中指定sharding key，那么GSM/shard director將請求連接到對應的一個shard
參數含義：
app_schema – 是業務用戶，
(host=gsm1)(port=1522) – 是GSM/shard director 監聽地址
service_name=oltp_rw_srvc.shdb.oradbcloud – 是前面創建的全局service
[oracle@db01 ~]$ sqlplus app_schema/oracle@'(description=(address=(protocol=tcp)(host=192.168.2.226)(port=1522)) (connect_data=(service_name=oltp_rw_srvc.shdb.oradbcloud)(region=region1)(SHARDING_KEY=chenjch)))'
SQL> select db_unique_name from v$database;
DB_UNIQUE_NAME
------------------------------
sh2
--- 插入數據
SQL> INSERT INTO Customers (CustId, FirstName, LastName, CustProfile,
Class, Geo, Passwd) VALUES ('chenjch', 'chen', 'jch',
NULL, 'Gold', 'east', hextoraw('8d1c00e')); 2 3
1 row created.
SQL> commit;
Commit complete.
SQL> select custid, FirstName, LastName, class, geo from customers where custid = 'chenjch';
CUSTID FIRSTNAME LASTNAME CLASS GEO
-------------------- --------------- --------------- ---------- --------
chenjch chen jch Gold east
--查詢SHARDING_KEY=tom.david，連接到sd2：
[oracle@db01 ~]$ sqlplus app_schema/oracle@'(description=(address=(protocol=tcp)(host=db01)(port=1522))
(connect_data=(service_name=oltp_rw_srvc.shdb.oradbcloud)(region=region1)(SHARDING_KEY=ch.jc)))'
SQL> select db_unique_name from v$database;
DB_UNIQUE_NAME
------------------------------
sh3
SQL> INSERT INTO Customers (CustId, FirstName, LastName, CustProfile,
Class, Geo, Passwd) VALUES ('chjc', 'ch', 'jc',
NULL, 'Gold', 'east', hextoraw('8d1c00a'));
1 row created.
SQL> commit;
Commit complete.
--- 查詢數據
---catalog本地查詢
SQL> conn app_schema/oracle
SQL> column custid format a20
SQL> column firstname format a15
SQL> column lastname format a15
SQL> select custid, FirstName, LastName, class, geo from customers;
CUSTID FIRSTNAME LASTNAME CLASS GEO
-------------------- --------------- --------------- ---------- --------
chenjch chen jch Gold east
chjc ch jc Gold east
---sh2本地查詢
SQL> conn app_schema/oracle
Connected.
SQL> column custid format a20
SQL> column firstname format a15
SQL> column lastname format a15
SQL> select custid, FirstName, LastName, class, geo from customers;
CUSTID FIRSTNAME LASTNAME CLASS GEO
-------------------- --------------- --------------- ---------- --------
chenjch chen jch Gold east
---sh3本地查詢
SQL> conn app_schema/oracle
Connected.
SQL> column custid format a20
SQL> column firstname format a15
SQL> column lastname format a15
SQL> select custid, FirstName, LastName, class, geo from customers;
CUSTID FIRSTNAME LASTNAME CLASS GEO
-------------------- --------------- --------------- ---------- --------
chjc ch jc Gold east
總結：
1
、每個shard節點只能查詢到本地節點shard table的數據
2
、在catalog數據庫可以查詢shard節點匯聚后的數據、
3
、客戶端可指定sharding key的方式，定向獲取數據
如果在連接串中指定sharding key，那么GSM/shard director將請求連接到對應的一個shard。
如果沒有指定sharding key，那么session和coordinator database (shard catalog)建立連接,然后再分別到需要(prund)的shard中查詢，最后再整合。
優化器判斷訪問一個shard還是訪問多個shard。
--鏈接到catalog數據庫查詢
[oracle@db01 ~]$ sqlplus app_schema/oracle@db01:1521/GDS\$CATALOG.oradbcloud
SQL> conn app_schema/oracle
Connected.
SQL> set termout on
SQL> set linesize 120
SQL> set echo on
SQL> column firstname format a20
SQL> column lastname format a20
SQL> explain plan for SELECT FirstName,LastName, geo, class FROM Customers
WHERE CustId in ('Scott.Tiger@x.bogus', 'Mary.Parker@x.bogus') AND class != 'free' ORDER
BY geo, class;
--- 參考
https://blog.csdn.net/kiral07/article/details/86923508#Shard_Catalog_Database_63
(1)catalog數據庫端,關閉director
GDSCTL>stop gsm -gsm SHARDDIRECTOR1
GSM 已成功停止
(2)關閉catalog監聽
[oracle@db01 ~]$ lsnrctl stop
(3)關閉catalog數據庫
SQL> shutdown immediate
(4) shard節點1和節點2關閉agent
[oracle@db02 trace]$ schagent -stop
[oracle@db03 trace]$ schagent -stop
(5) shard節點1和節點2關閉監控
[oracle@db02 trace]$ lsnrctl stop
[oracle@db03 trace]$ lsnrctl stop
(6) shard節點1和節點2關閉數據庫
SQL> shutdown immediate
(1)catalog端啟動數據庫和監聽
SQL> startup
[oracle@db01 ~]$ lsnrctl start
(2) 所有shard節點啟動數據庫，監聽，代理
SQL> startup
[oracle@db02 ~]$ lsnrctl start
[oracle@db03 ~]$ lsnrctl start
(3) catalog端啟動director
GDSCTL>start gsm -gsm SHARDDIRECTOR1
GSM 已成功啟動
(4)檢查
GDSCTL>connect mygds/oracle
已建立目錄連接
GDSCTL>config shard
名稱 分片組 狀態 狀態 區域 可用性
-- --- -- -- -- ---
sh2 primary_shardgroup 確定 DDL 錯誤 region1 ONLINE
sh3 primary_shardgroup 確定 DDL 錯誤 region1 ONLINE
GDSCTL>config service
名稱 網絡名稱 池 已啟動 全部首選
-- ---- - --- ----
oltp_rw_srvc oltp_rw_srvc.shdb.oradbcloud shdb 是 是
GDSCTL>databases
數據庫: "sh2" 注冊時間: Y 狀態: 確定 ONS: N。角色: PRIMARY 實例: 1 區域: region1
服務: "oltp_rw_srvc" 全局啟動: Y 啟動時間: Y
掃描: Y 啟用: Y 首選: Y
已注冊的實例:
shdb%1
數據庫: "sh3" 注冊時間: Y 狀態: 確定 ONS: N。角色: PRIMARY 實例: 1 區域: region1
服務: "oltp_rw_srvc" 全局啟動: Y 啟動時間: Y
掃描: Y 啟用: Y 首選: Y
已注冊的實例:
shdb%11
GDSCTL>create shard -shardgroup primary_shardgroup -destination db02 -credential cre_reg1 -sys_password oracle
GSM-45029: SQL 錯誤
ORA-03710: 目標中不存在目錄或者無法寫入目錄: $ORACLE_BASE/fast_recovery_area
ORA-06512: 在 "GSMADMIN_INTERNAL.DBMS_GSM_POOLADMIN", line 7469
ORA-06512: 在 "SYS.DBMS_SYS_ERROR", line 86
ORA-06512: 在 "GSMADMIN_INTERNAL.DBMS_GSM_POOLADMIN", line 4978
ORA-27436: 調度程序代理操作失敗, 顯示消息: Agent Error: /u01/app/oracle/fast_recovery_area/shard_check.txt (No such file or directory)
ORA-06512: 在 "SYS.DBMS_ISCHED", line 3638
ORA-06512: 在 "SYS.DBMS_ISCHED", line 10444
ORA-06512: 在 "SYS.DBMS_ISCHED", line 5344
ORA-06512: 在 "SYS.DBMS_ISCHED", line 10439
ORA-06512: 在 "SYS.DBMS_ISCHED", line 3631
ORA-06512: 在 "SYS.DBMS_SCHEDULER", line 2414
ORA-06512: 在 "GSMADMIN_INTERNAL.DBMS_GSM_POOLADMIN", line 164
ORA-27436: 調度程序代理操作失敗, 顯示消息: Agent Error: /u01/app/oracle/fast_recovery_area/shard_check.txt (No such file or directory)
ORA-06512: 在 "SYS.DBMS_ISCHED", line 3638
ORA-06512: 在 "SYS.DBMS_ISCHED", line 10444
ORA-06512: 在 "SYS.DBMS_ISCHED", line 5344
ORA-06512: 在 "SYS.DBMS_ISCHED", line 10439
ORA-06512: 在 "SYS.DBMS_ISCHED", line 3631
ORA-06512: 在 "SYS.DBMS_SCHEDULER", line 2414
ORA-06512: 在 "GSMADMIN_INTERNAL.DBMS_GSM_POOLADMIN", line 157
ORA-06512: 在 "GSMADMIN_INTERNAL.DBMS_GSM_POOLADMIN", line 4965
ORA-06512: 在 line 1
--- 錯誤提示很明顯，沒有fast_recovery_area 目錄，需要提前在db02 和db03 服務器上創建fast_recovery_area 目錄
[oracle@db02 ~]$ mkdir $ORACLE_BASE/fast_recovery_area
錯誤原因很多，需要具體問題具體分析，一般需要用到如下命令和日志進行分析問題：
(1)recover shard -shard sh2
(2)config shard -shard sh2
(3)alert_sh2.log
[oracle@db02 trace]$ pwd
/u01/app/oracle/diag/rdbms/sh2/sh2/trace
[oracle@db02 trace]$ tail -f alert_sh2.log
例如：
GDSCTL>recover shard -shard sh2
GSM 錯誤:
primary_shardgroup sh2:ORA-01119: error in creating database file \'/u01/app/oracle/oradata/db01/products_tsp01.dbf\'
ORA-06512: at "SYS.EXECASUSER", line 44
ORA-27040: file create error, unable to create file
Linux-x86_64 Error: 2: No such file or directory
Additional information: 1
ORA-06512: at "SYS.DBMS_SYS_SQL", line 1529
ORA-06512: at "SYS.EXECASUSER", line 31
ORA-06512: at "GSMADMIN_INTERNAL.EXECUTEDDL", line 70
ORA-06512: at line 1 (ngsmoci_execute)
GDSCTL>recover shard -shard sh2
GSM 錯誤:
primary_shardgroup sh2:ORA-03715: invalid user APP_SCHEMA
ORA-06512: at "SYS.DBMS_SYS_ERROR", line 86
ORA-06512: at "SYS.EXECASUSER", line 22
ORA-06512: at "GSMADMIN_INTERNAL.EXECUTEDDL", line 70
ORA-06512: at line 1 (ngsmoci_execute)
GDSCTL>config shard -shard sh3
名稱: sh3
分片組: primary_shardgroup
狀態: 確定
狀態: 已部署
區域: region1
連接字符串: db03:1521/sh3:dedicated
SCAN 地址:
ONS 遠程端口: 0
磁盤閾值, 毫秒: 20
CPU 閾值, %: 75
版本: 12.2.0.0
上次失敗的 DDL: drop tablespace TSP_SET_1 incl...
DDL 錯誤: ORA-03715: invalid user APP_SCHEMA
ORA-06512: at "SYS.DBMS_SYS_ERROR", line 86
ORA-06512: at "SYS.EXECASUSER", line 22
ORA-06512: at "GSMADMIN_INTERNAL.EXECUTEDDL", line 70
ORA-06512: at line 1 \(ngsmoci_execute\)
問題現象：
啟動GSM顯示實例已運行，停止GSM顯示異常或未運行；
排查原因：
查看告警日志
[oracle@db01 trace]$ pwd
/u01/app/oracle/diag/gsm/db01/sharddirector1/trace
[oracle@db01 trace]$ vi alert_sharddirector.log
問題原因：
創建sharddirector時指定的端口已經被其他應用占用，但是可以創建成功，后續啟動GSM就會失敗，刪除新建即可
GDSCTL> add gsm -gsm sharddirector1 -listener 1522 -pwd oracle -catalog db01:1521:db01 -region region1
《12c新特性-Oracle Sharding簡介》
https://blogs.oracle.com/database4cn/12c-oracle-sharding
《Sharded Database Deployment》
https://docs.oracle.com/en/database/oracle/oracle-database/12.2/admin/sharding-deployment.html#GUID-61B495D3-4482-47E2-937E-B7E03286F565
《Oracle 12C Sharding分片數據庫》
https://blog.csdn.net/kiral07/article/details/86923508#OracleSharding_Manage_80
《Oracle Sharding - Troubleshooting Tips and Techniques (文檔 ID 2180259.1)》
https://support.oracle.com/epmos/faces/SearchDocDisplay?_adf.ctrl-state=151coor2ou_9&_afrLoop=249031377428729
歡迎關注我的微信公眾號"IT小Chen"，共同學習，共同成長！！！
免責聲明：本站發布的內容（圖片、視頻和文字）以原創、轉載和分享為主，文章觀點不代表本網站立場，如果涉及侵權請聯系站長郵箱：is@yisu.com進行舉報，并提供相關證據，一經查實，將立刻刪除涉嫌侵權內容。
