Password less ssh for root user from compute node to all cell nodes

 Password less ssh for root user from compute node to all cell nodes:

Instead of copying the public key from the compute nodes to the autorizedkeys file of all the cell nodes; below command can be used for password less ssh for root user from the compute nodes to all the cell nodes.

Where /opt/oracle.SupportTools/onecommand/cell_group is the file where list of cellnodes is present.

#dcli -g /opt/oracle.SupportTools/onecommand/cell_group -l root -k -s '-o StrictHostKeyChecking=no'  

(Assuming ssh key is already present if not then run command #ssh-keygen -t rsa  )

To Verify password less is happening for all cells using uptime command.

#dcli -g /opt/oracle.SupportTools/onecommand/cell_group -l root uptime

===================================================================================

[root@dbadm01 onecommand]# dcli -g /opt/oracle.SupportTools/onecommand/cell_group -l root -k -s '-o StrictHostKeyChecking=no'

root@celadm01's password:

root@celadm02's password:

celadm01: ssh key added

celadm02: ssh key added

[root@dbadm01 onecommand]#

[root@dbadm01 onecommand]# dcli -g /opt/oracle.SupportTools/onecommand/cell_group -l root uptime

hcprd2celadm01: 12:57:01 up 165 days, 20:52,  0 users,  load average: 1.35, 1.91, 1.92

hcprd2celadm02: 12:57:01 up 165 days, 12:49,  0 users,  load average: 2.91, 2.15, 1.97

[root@dbadm01 onecommand]#

===================================================================================

Password less ssh for root user from compute node to all compute nodes:

Where /opt/oracle.SupportTools/onecommand/dbs_group is the file where list of compute nodes mentioned.

#dcli -g /opt/oracle.SupportTools/onecommand/dbs_group -l root -k -s '-o StrictHostKeyChecking=no'

#dcli -g /opt/oracle.SupportTools/onecommand/dbs_group -l root uptime

Regards,

Kiran Jadhav

Exadata cell node patching error_USB stayed back at old image

Exadata cell node patching error_USB stayed back at old image

 It seem USB stayed back at old image 20.1.9 though image was upgraded to 21.2.5; We can verify it in "imageinfo" command by taking login on problematic cell node.

#imageinfo

The Active image version and the Cell boot usb version should be same. if it different then the cell node patching takes time and it get fails with the error.

In this case, We need to rebuild USB using the below action plan which can be done online.

1. It may be required to stop the MS service to run this command.

#cellcli -e alter cell shutdown services MS

2. Recreate USB image with the following commands:

# cd /opt/oracle.SupportTools

# ./make_cellboot_usb -verbose -force

3. Remember to re-enable this once the make_cellboot_usb has completed.

#cellcli -e alter cell startup services MS

4. Once the execution is complete, validate configuration:

#cellcli -e alter cell validate configuration

5. Cross-verify the version

#imageinfo     ---> Active image version and the Cell boot usb version will be same.

6. Run cell node prechecks command and check if it still gives error.

Regards,

Kiran B Jaadhav

How to find serial number of cell nodes - Exadata or SSC

 How to find serial number of cell nodes:

In Exadata box or solaris supercluster (SSC) we may have multiple storage cell nodes attached. 

If we have 10-12 storage cells nodes then instead of login to each and every cells we can find out the serial number by using below command:

1. In Solaris super cluster:

#dcli -g /opt/oracle.supercluster/bin/cell_group -l root "dmidecode -s system-serial-number"

where # cat /opt/oracle.supercluster/bin/cell_group  --> will list number of cell nodes attached to the SSC machine

2. In Exadata servers:

#dcli -g /opt/oracle.SupportTools/onecommand/cell_group -l root "dmidecode -s system-serial-number"

where # cat /opt/oracle.SupportTools/onecommand/cell_group  --> will list number of cell nodes attached to the Exadata machine

Thank U

- Kiiran B Jaadhav