HP UNIX Oracle RAC node one disk corruption, node repair

Service guide for Oracle 10g RAC as a cluster-based software
Related directory information:
Ora_crs_home= $ORACLE _base/10.2/crs
OCR information:/dev/vgdata/rrac_ocr_1 640 ROOT:DBA
One of the nodes, two disk damage, from another node took a piece of hard disk to do mirror, change the hostname, IP, found that the database CRS can not start;

One check CRS log
No log generated, proving that it did not initiate operations on CRS
Two try to manually start Crsctl start CRS, error, CRSCTL check CRS process, found unable to communicate, check OS process, Ps-ef|grep–i D.bin, no valuable information, enter/SBIN/INIT.D start

Attempting to start CRS stackFailure at scls_scr_create with code 1Internal Error Information:  Category: 1234  Operation: scls_scr_create  Location: mkdir  Other: Unable to make user dir  Dep: 2

sh -x /sbin/iinit.d start
Check the operating system log:

/var/adm/messages/sbin/init.d/init.cssd[537]: /var/opt/oracle/scls_scr/rqtmsdb2/root/cssrun: Cannot create the specified file.Startup will be queued to init within 30 seconds.

This error is more meaningful, there is a file does not exist, then check the directory, does not exist, manually established;
Three again start CRS

crsctl start crs crsctl check crsCSS appears healthyCRS appears healthyEVM appears healthy

CRS,EVMD,CSSD started, but listen, the database fails to start
Four can actually guess the reason behind, because it is mirror, all and 1 of the node information is changed to 2, the back is more cumbersome
Check Alert_jkweb2.log:
Syslog:cluster ready Services waiting on dependencies. Diagnostics In/tmp/crsctl
Modify the log file directory under $ora_crs_home, start CRS again, but listen, the database still cannot start.
Check the listening parameters file: $ORACLE _home/network/admin/listener.ora
Manually modify the listening parameters file, database password file, parameter file
Manually pull up the database;
Five restart the host, CRS can not start, check the service guide:
cmrunnode -v
found that the PKG1 did not start, manually pull the package,
cmrunpkg –v –n jkweb1 pkg1
Check:

cmviewcl检查crs状态:cd $ORA_CRS_HOME/bin./crs_stat –t –v

Check the database status:
srvctl status database –d db_name
Everything is normal, work off;
Sometimes it feels like a service guide is a good thing;

HP UNIX Oracle RAC node one disk corruption, node repair