11g R2 RAC: ADD A NODE

11g R2 RAC has extremely simplified the process of adding a node to the cluster as compared to 10g RAC or 11g R1 RAC.  To know the  procedure to add a node in 10g RAC or 11g R1 RAC pls visit click here.
Now, after the introduction of SCAN and GPNP, we need to follow very simple steps.
Current scenario:
I have two nodes in the cluster presently.
Host names :
- host01.example.com
- host02.example.com
Node to be added : 
- host03.example.com
Procedure:
————————————
 Prepare the machine for third node
————————————
 — Set kernel parameters
 — Install required rpm’s
 — Create users/groups
  — configure oracleasm
root@host03#oracleasm configure -i
                       oracleasm exit
                       oracleasm init
                       oracleasm scandisks
                       oracleasm listdisks
              all asm disks will be listed
– Configure ssh connectivity for grid user among all 3 nodes –
   –  On  node3 as grid user
[grid@host03 .ssh]$ssh-keygen -t rsa
 
                             ssh-keygen -t dsa
 
                             cd /home/grid/.ssh
                              cat *.pub > host03
 
                             scp host03 host01:/home/grid/.ssh/
[grid@host03 .ssh] $ssh host01 
                           – Enter password
 [grid@host01 ~]$cd /home/grid/.ssh
                          
                          cat host03 >> authorized_keys
 
                         scp authorized_keys host02:/home/grid/.ssh/
                         scp authorized_keys host03:/home/grid/.ssh/
– Test ssh connectivity on all 3 nodes as grid user –
– run following  on all 3 nodes twice as grid user–
echo ssh host01 hostname >> a.sh
echo ssh host02 hostname >> a.sh
echo ssh host03 hostname >> a.sh
echo ssh host01-priv hostname >> a.sh
echo ssh host02-priv hostname >> a.sh
echo ssh host03-priv hostname  >> a.sh
 
chmod +x a.sh
 
./a.sh
– Run cluster verify to check that host03 can be added as node –
grid host01# cluvfy stage -pre crsinst -n host03 -verbose
– if time synchronization problem, restart ntpd service on each node
– Error grid is not a member of dba group – ignore
grid@host01 ~]$. oraenv –+ASM1
[grid@host01 ~]$ cd /u01/app/11.2.0/grid/oui/bin
– Add node 
[grid@host01 bin]$./addNode.sh -silent “CLUSTER_NEW_NODES={host03}” “CLUSTER_NEW_VIRTUAL_HOSTNAMES={host03-vip}”
– Execute oraInstroot.sh and root.sh on node3  as root –
[root@host03]#/u01/app/oraInventory/oraInstroot.sh
                         /u01/app/11.2.0/grid/root.sh
– check from host01 that node has been added –
host01 grid > crsctl stat res -t
– Start any resources if they are not up already –
host01 grid > crsctl start resource <resource name>

After you have added a node, in case you want to clone database home on that node, please click here.

References:

———————–

10 thoughts on “11g R2 RAC: ADD A NODE

  1. bro i have running ./addnode.sh script on oracle user connect to grid user [oracle@rac3]$cd $GRID_HOME/oui/bin then run script ./addnode .sh but its takes 10 and more why its take 10hrs or more time

  2. Also, we need to update /etc/hosts with new hostname, private and virtual.. Then copy the same file to other nodes..

    1. Hi Gyanendra

      You are correct. We need to update DNS / /etc/hosts with the new hostname. In my set up I was using DNS in which entry for host03 was already there.

      Regards
      Anju Garg

  3. Hi I am getting below error but I see /et6c/resolv.conf on both nodes havce same entry and I am not using DNS server

    File “/etc/resolv.conf” does not have both domain and search entries defined
    domain entry in file “/etc/resolv.conf” is consistent across nodes
    search entry in file “/etc/resolv.conf” is consistent across nodes
    All nodes have one search entry defined in file “/etc/resolv.conf”
    PRVF-5636 : The DNS response time for an unreachable node exceeded “15000” ms on following nodes: oratst01,oratst03

    File “/etc/resolv.conf” is not consistent across nodes

    Checking VIP configuration.
    Checking VIP Subnet configuration.
    Check for VIP Subnet configuration passed.
    Checking VIP reachability
    Check for VIP reachability passed.

    Pre-check for node addition was unsuccessful on all the nodes.

  4. Hi Anju
    when adding node do we need to install Grid and RDBMS home ? As per your blog not installing any binary . How does addnode.sh work ? Please clarify

    Regards
    Basheer

    1. Hi Basheer,

      addnode.sh is available in both grid home and database home. It copies binaries for the respective home to the new node and also installs the same.

      regards
      Anju

Your comments and suggestions are welcome!