如何在11.2集群中添加/删除资源？

最新推荐文章于 2022-04-04 03:24:08 发布

orion61

最新推荐文章于 2022-04-04 03:24:08 发布

阅读量2k

点赞数

分类专栏： Oracle Rac

Oracle Rac 专栏收录该内容

154 篇文章 0 订阅

订阅专栏

http://www.tuicool.com/articles/Z7RZRn

在11.2，数据库、监听、VIP、ASM、磁盘组等都属于GI所管理的资源，有的时候需要添加、删除、修改资源属性等，都是使用crsctl命令来操作的。
注意在集群中添加和删除资源只是将它从OCR中注册或者移除，并不是真正的将这个资源删掉。比如diskgroup这个资源，用crsctl add/delete并不会把这个磁盘组删除。

     下面列出了GI所管理的资源：
-bash-3.00$ crsctl stat res -t
--------------------------------------------------------------------------------
NAME           TARGET STATE        SERVER                   STATE_DETAILS
--------------------------------------------------------------------------------
Local Resources
--------------------------------------------------------------------------------
ora.DATA.dg
               ONLINE ONLINE       rac1
               ONLINE ONLINE       rac2
ora.LISTENER.lsnr
               ONLINE ONLINE       rac1
               ONLINE OFFLINE      rac2
ora.OCR_VOTE.dg
               ONLINE ONLINE       rac1
               ONLINE ONLINE       rac2
ora.RECO.dg
               ONLINE ONLINE       rac1
               ONLINE ONLINE       rac2
ora.asm
               ONLINE ONLINE       rac1                 Started
               ONLINE ONLINE       rac2                 Started
ora.gsd
               ONLINE OFFLINE      rac1
               ONLINE OFFLINE      rac2
ora.net1.network
               ONLINE ONLINE       rac1
               ONLINE OFFLINE      rac2
ora.ons
               ONLINE ONLINE       rac1
               ONLINE OFFLINE      rac2
--------------------------------------------------------------------------------
Cluster Resources
--------------------------------------------------------------------------------
ora.LISTENER_SCAN1.lsnr
      1        ONLINE ONLINE       rac1
ora.cvu
      1        ONLINE ONLINE       rac1
ora.rac1.vip
      1        ONLINE ONLINE       rac1
ora.rac2.vip
      1        ONLINE INTERMEDIATE rac1                 FAILED OVER
ora.oc4j
      1        ONLINE ONLINE       rac1
ora.rac11g2.db
      1        ONLINE ONLINE       rac1                 Open
      2        ONLINE ONLINE       rac2                 Open
ora.scan1.vip
      1        ONLINE ONLINE       rac1

官方文档中有添加、删除、修改资源的详细命令和解释：
http://docs.oracle.com/cd/E11882_01/rac.112/e16794/crsref.htm#CHDFEFBJ
Oracle Clusterware Administration and Deployment Guide
11g Release 2 (11.2)
E16794-17
添加资源：
crsctl add resource resource_name -type resource_type [-file file_path |
-attr "attribute_name=attribute_value,attribute_name=attribute_value,..."]
[-i] [-f]
resource_name A short, descriptive name for the resource. ( resource_name: 资源名）
-type resource_type The type of resource that you are adding preceded by the -type flag.（-type 资源类型）
-file file_path Path name (either absolute or relative) for a text file containing line-separated attribute name-value pairs that define the resource.
-attr "attribute_name= attribute_value （-attr: 资源属性以及对应的值，不同类型的资源对应的属性不同）
-i If you specify -i, then the command returns an error if processing this command requires waiting for Oracle Clusterware to unlock the resource or its dependents. Sometimes, Oracle Clusterware locks resources or other objects to prevent commands from interfering with each other.
-f Use the force option:（-f : 强制添加，忽略依赖性）
    To add a resource that has dependencies on other resources that do not yet exist. The force option overrides checks that would prevent a command from being completed.
    To add a resource if the resource has hard dependencies on other resources and the owner of the resources does not execute permissions on one or more of the dependencies. If you do not specify the force option in this case, an error displays.
    To add resources of application type because you may need to move servers into the Generic server pool. If the servers currently host resources that must be stopped, then the force option is required
删除资源：
crsctl delete resource resource_name [-i] [-f]
resource_name Specify the name of the resource you want to remove.
-i If you specify -i, then the command returns an error if processing this command requires waiting for Oracle Clusterware to unlock the resource or its dependents. Sometimes, Oracle Clusterware locks resources or other objects to prevent commands from interfering with each other.
-f Use the force option to remove either running resources, or remove this resource even though other resources have a hard dependency on it.（-f 强制删除，即使这个资源在运行或者有其他资源依赖于它）
修改资源：
crsctl modify resource resource_name -attr "attribute_name=attribute_value"
[-i] [-f] [-delete]
resource_name The name of the resource you want to modify.
-attr "attribute_name=attribute_value"
-i If you specify -i, then the command returns an error if processing this command requires waiting for Oracle Clusterware to unlock the resource or its dependents. Sometimes, Oracle Clusterware locks resources or other objects to prevent commands from interfering with each other.
-f Use the -f option when:
    The resource has a hard dependency on a non-existing resource
    The owner of the resource does not have execute permissions on one or more hard dependencies
    The modification results in servers being moved into the Generic pool and resources being stopped or relocated to accomplish the server move
-delete If you specify the -delete option, then Oracle Clusterware deletes the named attribute.
下面是在11.2.0.3上删除和添加diskgroup ora.RECO.dg的例子：

主要用到的命令：
su - grid

1. 停止依赖于磁盘组ora.RECO.dg的所有数据库:
$ srvctl stop database -d <db_name>

2. 查看ora.RECO.dg的配置信息：
-bash-3.00$ crsctl stat res ora.RECO.dg -p
NAME=ora.RECO.dg
TYPE=ora.diskgroup.type
ACL=owner:grid:rwx,pgrp:oinstall:rwx,other::r-- <===这个信息在添加资源的使用需要
...
VERSION=11.2.0.3.0

3. 停止ora.RECO.dg：
-bash-3.00$ crsctl stop res ora.RECO.dg -f

4. 从集群中移除ora.RECO.dg：
-bash-3.00$ crsctl delete res ora.RECO.dg -f
5. 将ora.RECO.dg添加到集群中：
-bash-3.00$ crsctl add res ora.RECO.dg -type ora.diskgroup.type -attr "ACL='owner:grid:rwx,pgrp:oinstall:rwx,other::r-
-',AUTO_START=always,VERSION=11.2.0.3.0" -i
6. 检查资源的配置信息是正确的：
-bash-3.00$ crsctl stat res ora.RECO.dg -p

7. 将这个资源启动：
-bash-3.00$ crsctl start res ora.RECO.dg
-bash-3.00$ crsctl stat res -t

8. 启动所有依赖于这个磁盘组的数据库：
$ srvctl start database -d <db_name>

具体的测试过程及输出：
su - grid

1. 停止依赖于磁盘组ora.RECO.dg的所有数据库:
$ srvctl stop database -d RAC11G2

2. 查看ora.RECO.dg的配置信息：

-bash-3.00$ crsctl stat res ora.RECO.dg -p
NAME=ora.RECO.dg
TYPE=ora.diskgroup.type
ACL=owner:grid:rwx,pgrp:oinstall:rwx,other::r--
ACTION_FAILURE_TEMPLATE=
ACTION_SCRIPT=
AGENT_FILENAME=%CRS_HOME%/bin/oraagent%CRS_EXE_SUFFIX%
ALIAS_NAME=
AUTO_START=always
CHECK_INTERVAL=300
CHECK_TIMEOUT=30
DEFAULT_TEMPLATE=
DEGREE=1
DESCRIPTION=CRS resource type definition for ASM disk group resource
ENABLED=1
LOAD=1
LOGGING_LEVEL=1
NLS_LANG=
NOT_RESTARTING_TEMPLATE=
OFFLINE_CHECK_INTERVAL=0
PROFILE_CHANGE_TEMPLATE=
RESTART_ATTEMPTS=5
SCRIPT_TIMEOUT=60
START_DEPENDENCIES=hard(ora.asm) pullup(ora.asm)
START_TIMEOUT=900
STATE_CHANGE_TEMPLATE=
STOP_DEPENDENCIES=hard(intermediate:ora.asm)
STOP_TIMEOUT=180
TYPE_VERSION=1.2
UPTIME_THRESHOLD=1d
USR_ORA_ENV=
USR_ORA_OPI=false
USR_ORA_STOP_MODE=
VERSION=11.2.0.3.0

3. 停止ora.RECO.dg：
-bash-3.00$ crsctl stop res ora.RECO.dg -f
CRS-2673: Attempting to stop 'ora.RECO.dg' on 'rac2'
CRS-2673: Attempting to stop 'ora.RECO.dg' on 'rac1'
CRS-2677: Stop of 'ora.RECO.dg' on 'rac2' succeeded
CRS-2677: Stop of 'ora.RECO.dg' on 'rac1' succeeded

4. 从集群中移除ora.RECO.dg：
-bash-3.00$ crsctl delete res ora.RECO.dg   <==如果有不加-f, 当有资源依赖于这个磁盘组时会报错
CRS-2730: Resource 'ora.rac11g2.db' depends on resource 'ora.RECO.dg'
CRS-4000: Command Delete failed, or completed with errors.

-bash-3.00$ crsctl delete res ora.RECO.dg -f      <=====需要使用-f来强制删除

5. 将ora.RECO.dg添加到集群中：
-bash-3.00$ crsctl add res ora.RECO.dg -type ora.diskgroup.type -attr "ACL='owner:grid:rwx,pgrp:oinstall:rwx,other::r-

-',AUTO_START=always,VERSION=11.2.0.3.0" -i

6. 检查资源的配置信息是正确的：

-bash-3.00$ crsctl stat res ora.RECO.dg -p
NAME=ora.RECO.dg
TYPE=ora.diskgroup.type
ACL=owner:grid:rwx,pgrp:oinstall:rwx,other::r--
ACTION_FAILURE_TEMPLATE=
ACTION_SCRIPT=
AGENT_FILENAME=%CRS_HOME%/bin/oraagent%CRS_EXE_SUFFIX%
ALIAS_NAME=
AUTO_START=always
CHECK_INTERVAL=300
CHECK_TIMEOUT=30
DEFAULT_TEMPLATE=
DEGREE=1
DESCRIPTION=CRS resource type definition for ASM disk group resource
ENABLED=1
LOAD=1
LOGGING_LEVEL=1
NLS_LANG=
NOT_RESTARTING_TEMPLATE=
OFFLINE_CHECK_INTERVAL=0
PROFILE_CHANGE_TEMPLATE=
RESTART_ATTEMPTS=5
SCRIPT_TIMEOUT=60
START_DEPENDENCIES=hard(ora.asm) pullup(ora.asm)
START_TIMEOUT=900
STATE_CHANGE_TEMPLATE=
STOP_DEPENDENCIES=hard(intermediate:ora.asm)
STOP_TIMEOUT=180
TYPE_VERSION=1.2
UPTIME_THRESHOLD=1d
USR_ORA_ENV=
USR_ORA_OPI=false
USR_ORA_STOP_MODE=
VERSION=11.2.0.3.0

7. 将这个资源启动：
-bash-3.00$ crsctl start res ora.RECO.dg
CRS-2672: Attempting to start 'ora.RECO.dg' on 'rac1'
CRS-2672: Attempting to start 'ora.RECO.dg' on 'rac2'
CRS-2676: Start of 'ora.RECO.dg' on 'rac1' succeeded
CRS-2676: Start of 'ora.RECO.dg' on 'rac2' succeeded

-bash-3.00$ crsctl stat res -t
--------------------------------------------------------------------------------
NAME           TARGET STATE        SERVER                   STATE_DETAILS       Local Resources
--------------------------------------------------------------------------------
ora.RECO.dg
              ONLINE ONLINE       rac1
              ONLINE ONLINE       rac2

8.启动所有依赖于这个磁盘组的数据库：
$ srvctl start database -d RAC11G2