Archives October 13, 2020

prvf-4123-inconsistent-user-ids-found-for-user-oracle

Solution:

User id of oracle on node 1 mismatch with node 2.

I tried to add id 500 for oracle user on both nodes.I find 500 userid on node 1 and 501 userid on node 2!!

[oracle@RACFLEX1 grid]$ id oracle
uid=500(oracle) gid=500(oinstall) groups=500(oinstall),501(dba),502(asmadmin),503(asmdba),504(asmoper)

[root@RACFLEX2 oracle]# id oracle
uid=501(oracle) gid=500(oinstall) groups=500(oinstall),501(dba),502(asmadmin),503(asmdba),504(asmoper)

I try to modify the id on node 2 as 500 but couldnot do that.

[root@RACFLEX2 oracle]# usermod -u 500 -g oinstall -G dba,asmdba,asmadmin,asmoper oracle                                                                                                             
usermod: UID '500' already exists
[root@RACFLEX2 oracle]# userdel oracle
userdel: user oracle is currently used by process 4626
[root@RACFLEX2 oracle]# kill -9 4626

Some other user is using the same id .I have to figure out which user?

[root@RACFLEX2 ~]# cat /etc/passwd|grep 500 
grid:x:500:500::/home/grid:/bin/bash
oracle:x:501:500::/home/oracle:/bin/bash

I check password file and found out grid user using the same id.But i dont want the grid user anymore.So i deleted grid user and readd oracle user with id 500

[root@RACFLEX2 ~]# userdel grid
[root@RACFLEX2 ~]# usermod -u 500 -g oinstall -G dba,asmdba,asmadmin,asmoper ora                                                                                                             cle
[root@RACFLEX2 ~]# id oracle
uid=500(oracle) gid=500(oinstall) groups=500(oinstall),501(dba),502(asmadmin),50                                                                                                             3(asmdba),504(asmoper)
[INS-30515] OCR disks size and count by redundancy

In oracle RAC,OCR file is accessed by clusterware to start the crsd agents and processes.

Have you wondered,how much size and how many disks should we allocate according to different redundancy??

External redundancy:

for external redundancy, we should atleast have diskspace of 5.5GB and 1 disk required for redundancy in OCR diskgroup

Normal redundancy:

For normal redundancy, we should have atleast 11GB diskspace and 3 disks required for redundancy in OCR diskgroup

High redundancy:

For high redundancy, we should have atleast 16.5GB diskspace and 5 disks required for redundancy in OCR diskgroup !!

keep in mind that for 12c grid version,we require below for OCR diskgroup configuration

external – minimum 5.5G disk space with 1 disk or more

normal – minimum 11G disk space with 3 disk or more

high – minimum 16.5G disk space with 5 disk or more

error: %pre() scriptlet failed, exit status 4error: install: %pre scriptlet failed (2), skipping cell-11.2.3.2.1_LINUX.X64_130109-1

Solution:

[root@EXDBADM001 cellbits]# rpm -qa|grep jdk

jdk-1.5.0_15-fcs.x86_64 <<<<<<<

There is a mismatch in format of rpm installation package

Create a macros file on the configuration directory and specify the correct format!! and retry rpm install with nodeps option. 
eg:rpm -Uvh xxx.rpm --nodeps

[root@EXDBADM001 cellbits]# cd ~/.

[root@EXDBADM001 cellbits]# cat ~/.rpmmacros
%_query_all_fmt %%{name}-%%{version}-%%{release}
11.2.0.4 runInstaller: [INS-06006] Passwordless SSH connectivity not set up between the following nodes(s)

Solution:

Edit /etc/hosts file and specify hostname of both nodes in uppercase and bounce the RAC nodes.
eg: RAC1 should be rac1