Table of Contents
Methode 1 : Using rootdelete.sh and rootdeinstall.sh on all rac nodes
# cd /u01/app/oracle/product/crs/install
# ./rootdelete.sh
# ./rootdeinstall.sh
# rm -rf /var/tmp/.oracle
Now rerun root.sh
on ract1
# /u01/app/oracle/product/crs/root.sh
Later run on ract2
# /u01/app/oracle/product/crs/root.sh
Methode 2: Manually deletion
1) Try to stop nodeapps on all cluster nodes [root@ract1 install]# srvctl stop nodeapps -n ract1 PRKH-1010 : Unable to communicate with CRS services. [OCR Error(Native: prsr_initCLSS:[21])] [root@ract2 ~]# srvctl stop nodeapps -n ract2 PRKH-1010 : Unable to communicate with CRS services. [OCR Error(Native: prsr_initCLSS:[21])] 2) Cleanup files on all cluster nodes rm /etc/oracle/* rm -f /etc/init.d/init.cssd rm -f /etc/init.d/init.crs rm -f /etc/init.d/init.crsd rm -f /etc/init.d/init.evmd rm -f /etc/rc2.d/K96init.crs rm -f /etc/rc2.d/S96init.crs rm -f /etc/rc3.d/K96init.crs rm -f /etc/rc3.d/S96init.crs rm -f /etc/rc5.d/K96init.crs rm -f /etc/rc5.d/S96init.crs rm -Rf /etc/oracle/scls_scr rm -f /etc/inittab.crs cp /etc/inittab.orig /etc/inittab 3) Check CRS processes and kill them on all cluster nodes [root@ract2 rules.d]# ps -ef | egrep 'css|crs|evm' root 4314 1 0 07:13 ? 00:00:00 /bin/sh /etc/init.d/init.evmd run root 4315 1 0 07:13 ? 00:00:00 /bin/sh /etc/init.d/init.cssd fatal root 4316 1 0 07:13 ? 00:00:00 /bin/sh /etc/init.d/init.crsd run root 4343 4315 0 07:13 ? 00:00:00 /bin/sh /etc/init.d/init.cssd startcheck root 4365 4314 0 07:13 ? 00:00:00 /bin/sh /etc/init.d/init.cssd startcheck root 4516 4316 0 07:13 ? 00:00:00 /bin/sh /etc/init.d/init.cssd startcheck root 16626 9372 0 08:57 pts/1 00:00:00 egrep css|crs|evm [root@ract2 rules.d]# kill -9 4314 4315 4316 4343 4365 4516 [root@ract2 rules.d]# ps -ef | egrep 'css|crs|evm' root 18063 9372 0 09:10 pts/1 00:00:00 egrep css|crs|evm 4) Cleanup /var/tmp/.oracle/ and /tmp/.oracle/ on all cluster nodes If there is no other Oracle software running (like listeners, DB's, etc...), you can remove the files in /var/tmp/.oracle or /tmp/.oracle. [root@ract2 ~]# rm -f /var/tmp/.oracle/* or rm -f /tmp/.oracle/* 5) Remove the ocr.loc Usually the ocr.loc can be found at /etc/oracle on all cluster nodes --> should already clean due to step 1 6) Clean out the OCR and Voting Files with dd commands - only on a single node [root@ract2 ~]# /bin/dd if=/dev/zero skip=25 bs=4k count=2560 of=/dev/raw/raw1 2560+0 records in 2560+0 records out 10485760 bytes (10 MB) copied, 0.248064 seconds, 42.3 MB/s [root@ract2 ~]# /bin/dd if=/dev/zero skip=25 bs=4k count=2560 of=/dev/raw/raw2 [root@ract2 ~]# /bin/dd if=/dev/zero skip=25 bs=4k count=2560 of=/dev/raw/raw3 [root@ract2 ~]# /bin/dd if=/dev/zero skip=25 bs=4k count=2560 of=/dev/raw/raw4 [root@ract2 ~]# /bin/dd if=/dev/zero skip=25 bs=4k count=2560 of=/dev/raw/raw5
Reference
- How to Proceed From a Failed 10g or 11.1 Oracle Clusterware (CRS) Installation (Doc ID 239998.1)