What's preconnect.svc in 11g RAC?

有网友反映在11.2的RAC中有一个名如*_preconnect.svc的资源一直处于OFFLINE状态，而其TARGET则为ONLINE状态，无法通过重启CRS或者手动start该资源来使之ONLINE。我们来具体看一下什么情况下回产生这种资源?:

[oracle@rh2 ~]$ srvctl add service -h

Adds a service configuration to the Oracle Clusterware.

Usage: srvctl add service -d  -s  {-r "" 
[-a ""] [-P {BASIC | NONE | PRECONNECT}] | -g  
[-c {UNIFORM | SINGLETON}] } [-k   ] [-l [PRIMARY][,PHYSICAL_STANDBY]
[,LOGICAL_STANDBY][,SNAPSHOT_STANDBY]] [-y {AUTOMATIC | MANUAL}] [-q {TRUE|FALSE}] 
[-x {TRUE|FALSE}] [-j {SHORT|LONG}] [-B {NONE|SERVICE_TIME|THROUGHPUT}] 
[-e {NONE|SESSION|SELECT}] [-m {NONE|BASIC}] [-z ] 
[-w ] [-t ] [-f]
    -d       Unique name for the database
    -s              Service name
    -r ""    Comma separated list of preferred instances
    -a ""    Comma separated list of available instances
    -g            Server pool name
    -c {UNIFORM | SINGLETON} Service runs on every active server in the 
server pool hosting this service (UNIFORM) or just one server (SINGLETON)
    -k              network number (default number is 1)
    -P {NONE | BASIC | PRECONNECT}        TAF policy specification
    -l                 Role of the service (primary, physical_standby, logical_standby, snapshot_standby)
    -y               Management policy for the service (AUTOMATIC or MANUAL)
    -e        Failover type (NONE, SESSION, or SELECT)
    -m      Failover method (NONE or BASIC)
    -w              Failover delay
    -z              Failover retries
    -t              Edition (or "" for empty edition value)
    -j   Connection Load Balancing Goal (SHORT or LONG). Default is LONG.
    -B      Runtime Load Balancing Goal (SERVICE_TIME, THROUGHPUT, or NONE)
    -x   Distributed Transaction Processing (TRUE or FALSE)
    -q  AQ HA notifications (TRUE or FALSE)
Usage: srvctl add service -d  -s  -u {-r "" | -a ""} [-f]
    -d       Unique name for the database
    -s              Service name
    -u                       Add a new instance to service configuration
    -r        Name of new preferred instance
    -a       Name of new available instance
    -f                       Force the add operation even though a listener is not configured for a network
    -h                       Print usage

[oracle@rh2 ~]$ srvctl add service -d PROD -s maclean -r "PROD1,PROD2" -P BASIC

[oracle@rh2 ~]$ crs_stat|grep maclean
NAME=ora.prod.maclean.svc
NAME=ora.prod.maclean_taf.svc

[oracle@rh2 ~]$ srvctl add service -d PROD -s maclean_pre -r "PROD1,PROD2" -P PRECONNECT
[oracle@rh2 ~]$ crs_stat|grep maclean_pre            
NAME=ora.prod.maclean_pre.svc
NAME=ora.prod.maclean_pre_preconnect.svc

/* 可以看到仅当创建的service使用preconnect的TAF policy时会附带创建{service_name}_preconnect.svc的服务 */

[oracle@rh2 ~]$ srvctl start service -d PROD -s maclean    
[oracle@rh2 ~]$ srvctl start service -d PROD -s maclean_pre

[oracle@rh2 ~]$ crs_stat |grep -A3 maclean
NAME=ora.prod.maclean.svc
TYPE=ora.service.type
TARGET=ONLINE
STATE=ONLINE on rh2
--
NAME=ora.prod.maclean_pre.svc
TYPE=ora.service.type
TARGET=ONLINE
STATE=ONLINE on rh2
--
NAME=ora.prod.maclean_pre_preconnect.svc
TYPE=ora.service.type
TARGET=ONLINE
STATE=OFFLINE
--
NAME=ora.prod.maclean_taf.svc
TYPE=ora.service.type
TARGET=OFFLINE
STATE=OFFLINE

[oracle@rh2 ~]$ srvctl start service -d prod -s maclean_pre_preconnect
PRCD-1084 : Failed to start service maclean_pre_preconnect
PRCR-1079 : Failed to start resource ora.prod.maclean_pre_preconnect.svc
CRS-2674: Start of 'ora.prod.maclean_pre_preconnect.svc' on 'rh2' failed
CRS-2674: Start of 'ora.prod.maclean_pre_preconnect.svc' on 'rh3' failed
CRS-2632: There are no more servers to try to place resource 
'ora.prod.maclean_pre_preconnect.svc' on that would satisfy its placement policy

/* 无法手动启动该preconnect.svc服务 */

[oracle@rh2 ~]$ crs_stat |grep preconnect
NAME=ora.prod.maclean_pre_preconnect.svc

[oracle@rh2 ~]$ srvctl remove service -d PROD -s maclean_pre_preconnect
PRCD-1107 : Removed service maclean_pre_preconnect but failed to remove its 
underlying server pool PROD_maclean_pre
PRCS-1012 : Failed to remove server pool PROD_maclean_pre
PRCR-1072 : Failed to unregister server pool ora.PROD_maclean_pre
CRS-2554: Server pool 'ora.PROD_maclean_pre' cannot be unregistered as 
it is referenced by resource 'ora.prod.maclean_pre.svc'

[oracle@rh2 ~]$ crs_stat |grep preconnec

/* 以上remove命令报错但该preconnect.svc服务还是被删除了，虽然我们并不推荐这样做 */

[oracle@rh2 admin]$ srvctl start service -d PROD -s maclean_pre

[oracle@rh2 admin]$ lsnrctl service

LSNRCTL for Linux: Version 11.2.0.2.0 - Production on 30-MAR-2011 18:55:43

Copyright (c) 1991, 2010, Oracle.  All rights reserved.

Connecting to (ADDRESS=(PROTOCOL=tcp)(HOST=)(PORT=1521))
Services Summary...
Service "+ASM" has 1 instance(s).
  Instance "+ASM1", status READY, has 1 handler(s) for this service...
    Handler(s):
      "DEDICATED" established:0 refused:0 state:ready
         LOCAL SERVER
Service "PROD" has 1 instance(s).
  Instance "PROD1", status READY, has 1 handler(s) for this service...
    Handler(s):
      "DEDICATED" established:0 refused:0 state:ready
         LOCAL SERVER
Service "PRODXDB" has 1 instance(s).
  Instance "PROD1", status READY, has 1 handler(s) for this service...
    Handler(s):
      "D000" established:0 refused:0 current:0 max:1022 state:ready
         DISPATCHER 
         (ADDRESS=(PROTOCOL=tcp)(HOST=rh2)(PORT=36196))
Service "maclean" has 1 instance(s).
  Instance "PROD1", status READY, has 1 handler(s) for this service...
    Handler(s):
      "DEDICATED" established:0 refused:0 state:ready
         LOCAL SERVER
Service "maclean_pre" has 1 instance(s).
  Instance "PROD1", status READY, has 1 handler(s) for this service...
    Handler(s):
      "DEDICATED" established:0 refused:0 state:ready
         LOCAL SERVER
The command completed successfully
                                                                                                                           
[oracle@rh2 admin]$ tnsping PROD_TAF

TNS Ping Utility for Linux: Version 11.2.0.2.0 - Production on 30-MAR-2011 18:54:38

Copyright (c) 1997, 2010, Oracle.  All rights reserved.

Used parameter files:

Used TNSNAMES adapter to resolve the alias
Attempting to contact (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP)
(HOST = rh-cluster-scan)(PORT = 1521)) (CONNECT_DATA = (SERVER = DEDICATED) 
(SERVICE_NAME = maclean_pre)))
OK (10 msec)


[oracle@rh2 admin]$ sqlplus  maclean/maclean@PROD_TAF

SQL> select instance_name,host_name from v$instance;

INSTANCE_NAME
----------------
HOST_NAME
----------------------------------------------------------------
PROD2
rh3.oracle.com

shutdown PROD2 instance:SQL> shutdown abort

SQL> select instance_name,host_name from v$instance;

INSTANCE_NAME
----------------
HOST_NAME
----------------------------------------------------------------
PROD1
rh2.oracle.com


/* 可以看到即使删除preconnect.svc服务也不会影响到TAF的正常使用，
    在生产环境中无需刻意去调试该服务，最好的应对方式是放任不管
*/

所以如果某天你看到你的crs_stat -t输出里有一条莫名的类似"ora....ect.svc ora....ce.type ONLINE OFFLINE"的记录，那么大可以忽略该资源的OFFLINE状态；这是正常现象，用不着紧张！