TCP.VALIDNODE_CHECKING导致crs无法启动
linux 三节点 rac，一个节点无法启动crs日志
2014-04-22 19:11:01.754: [ CRSMAIN] Checking compatibility...
2014-04-22 19:11:01.754: Static Version 11.2.0.3.0
2014-04-22 19:11:01.756: Daemon version: 11.2.0.3.0 Software version: 11.2.0.3.0
2014-04-22 19:11:01.757: Active Version from OCR:11.2.0.3.0
2014-04-22 19:11:01.757: Active Version and Software Version are same
2014-04-22 19:11:01.758: Active Version changed to 11.2.0.3.0
2014-04-22 19:11:01.758: [ CRSMAIN] Registering for mastership change events...
2014-04-22 19:11:01.758: th_reg_master_change: Master change callback registered
2014-04-22 19:11:01.758: a_reg_master_change: Registered master change callback
2014-04-22 19:11:01.758: th_not_master_change: Invoking master change callback. Master Inc
2014-04-22 19:11:01.758: a_reg_master_change: Notified master change
2014-04-22 19:11:01.758: [ CRSMAIN] Registering with boot group...
2014-04-22 19:11:01.762: [ CRSMAIN] Running upgrade config...
2014-04-22 19:11:01.782: [ CRSMAIN] Running path init...
2014-04-22 19:11:01.788: [ CLSE]clse_get_auth_loc: Returning default authloc: /oracle/11.2.0/grid/auth/crs/tdxtdb3
2014-04-22 19:11:01.788: [ CRSMAIN] Using Authorizer location: /oracle/11.2.0/grid/auth/crs/tdxtdb3
2014-04-22 19:11:01.788: [ CRSMAIN] Initialing cluclu context...
2014-04-22 19:11:01.797: clsclu_init: rc 0
2014-04-22 19:11:01.797: [ CRSMAIN] Getting CR Root...
2014-04-22 19:11:01.800: [ CRSMAIN] Initializing RTI
2014-04-22 19:11:01.800: [ CRSMAIN] Initializing staging area
2014-04-22 19:11:01.837: [ CLSE]clse_get_auth_loc: Returning default authloc: /oracle/11.2.0/grid/auth/crs/tdxtdb3
2014-04-22 19:11:01.837: [ CRSD] AuthLoc /oracle/11.2.0/grid/auth/crs/tdxtdb3
2014-04-22 19:11:01.837: [ CRSD] PE active version: 11.2.0.3.0
2014-04-22 19:11:01.837: [ CRSD] PE Engine: NEW
2014-04-22 19:11:01.837: [ CRSD] Using OCR batch ops : ENABLED
2014-04-22 19:11:01.837: [ CRSMAIN] Creating RTI lock info...
2014-04-22 19:11:01.837: [ CRSMAIN] Initializing EVMMgr
2014-04-22 19:11:01.841: [ CRSMAIN] Getting local nodename...
[ CLWAL]clsw_Initialize: OLR initlevel
2014-04-22 19:11:02.086: [ COMMCRS]clsclisten: Error listening on: (ADDRESS=(PROTOCOL=tcp)(HOST=192.168.1.79)(PORT=0))
2014-04-22 19:11:02.086: [ COMMCRS]clsclisten: op 65 failed, NSerr (12560, 0), transport: (583, 0, 0)
2014-04-22 19:11:02.086: [ CRSMAIN] Created alert : (:CRSD00133:) :Unable to get E2E port, error: IOException : clsclisten failed with ret 3
(File: caa_Socket.cpp, line: 525
2014-04-22 19:11:02.086: [ CRSD] CRSD exiting: Unable to get E2E port after 2nd attempt
2014-04-22 19:11:02.086: [ CRSD] Done.
分析grid的sqlnet.ora文件
more $GRID_HOME/network/admin/sqlnet.ora
# sqlnet.ora.tdxtdb3 Network Configuration File: /oracle/11.2.0/grid/network/adminlnet.ora.tdxtdb3
# Generated by Oracle configuration tools.
NAMES.DIRECTORY_PATH= (TNSNAMES, EZCONNECT)
ADR_BASE = /oracle/app
#TCP.VALIDNODE_CHECKING=yes
TCP.VALIDNODE_CHECKING=no
处理方法：注释掉TCP.VALIDNODE_CHECKING=no
参考文档
CTSS Daemon Aborting With "op 65 failed, NSerr (12560, 0), transport: (583, 0, 0)" (Doc ID 1087521.1)APPLIES TO:
Oracle Server - Enterprise Edition - Version: 11.2.0.1 and later
Information in this document applies to any platform.
SYMPTOMS
ora.ctssd fails to start
octssd.log
2010-04-14 10:14:03.592: [ CTSS]CTSS daemon aborting
2010-04-14 10:24:09.197: [ CTSS]Oracle Database CTSS Release 11.2.0.1.0 Production Copyright 2006, 2007 Oracle. All rights
reserved.
...
2010-04-14 10:24:09.641: [ CTSS]ctss_init: CTSS production mode
2010-04-14 10:24:09.641: [ CTSS]ctss_init: Env var CTSS_REBOOT is undefined or contains non-boolean value. Ignoring
CTSS_REBOOT.
2010-04-14 10:24:09.641: [ CTSS]sclsctss_gvss1: NTP default config file found
2010-04-14 10:24:09.641: [ CTSS]sclsctss_gvss8: Return and NTP status .
2010-04-14 10:24:09.641: [ CTSS]ctss_check_vendor_sw: Vendor time sync software is detected. status .
2010-04-14 10:24:09.641: [ CTSS]ctss_check_vendor_sw: Ctssd is switching to observer role
2010-04-14 10:24:09.643: [ CTSS]ctsscomm_init: The Socket name is [(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.0.2))]
2010-04-14 10:24:09.650: [ COMMCRS]clsclisten: Error listening on: (ADDRESS=(PROTOCOL=tcp)(HOST=192.168.0.2))
2010-04-14 10:24:09.650: [ COMMCRS]clsclisten: op 65 failed, NSerr (12560, 0), transport: (583, 0, 0)
2010-04-14 10:24:09.650: [ CTSS](:ctsscomm_init3:) clsclisten [(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.0.2))] failed . Return
.
2010-04-14 10:24:09.650: [ CTSS]ctss_main: COMM init failed
2010-04-14 10:24:09.650: [ CTSS]ctss_main: CTSS daemon aborting .
2010-04-14 10:24:09.650: [ CTSS]CTSS daemon aborting
CHANGES
TCP.VALIDNODE_CHECKING and TCP.INVITED_NODES have been set in $GRID_HOME/network/admin/sqlnet.ora
CAUSE
The error message is the same as in unpublished Bug 4621978
op 65 failed, NSerr (12560, 0), transport: (583, 0, 0)
According to the bug the problem is related to TCP.VALIDNODE_CHECKING and TCP.INVITED_NODES parameters.
SOLUTION
1. Remove TCP.VALIDNODE_CHECKING and TCP.INVITED_NODES from $GRID_HOME/network/admin/sqlnet.ora and restart Grid Infrastructure
or
2. Add the private interconnect IPs to TCP.INVITED_NODES and restart Grid Infrastructure
页:
[1]