Skip to main content

Manual Failover steps of DB nodes while on LVM

In order to facilitate manual failover below are the steps a sys admin should take when failing over the filesystems from one node to another when we have an LVM configured. NOTE: NODE02 uses ifcfg-eth5:0 for the listener VIP and NODE01 uses ifcfg-eth4:0 for the VIP. Please start these respectively.

On NODE02
1. Ensure the DB, DB listeners and the VIP has been stopped on the active node.
· Stop Oracle DB
· Stop Oracle Listener
· Stop the VIP Pointing to this DB Instance ( i.e. eth5:0)

# Ifdown eth5:0

2. Unmount all mount points related to PDB:

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
/dev/mapper/vg--oradata-lv.pdb01 40G 8.3G 30G 23% /pdb01 /dev/mapper/vg--oradata-lv.pdb02 493G 147G 321G 32% /pdb02/dev/mapper/vg--oradata-lv.pdb03 493G 133G 335G 29% /pdb03/dev/mapper/vg--oradata-lv.pdb_arch 197G 84G 104G 45% /pdb_arch/dev/mapper/vg--oradata-lv.dbdump 49G 180M 46G 1% /dbdump/pdb
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
# umount pdb* ( this unmounts all file systems with the name pdb..)

3. Check the status of LV on Active node (Node02).

# lvscan ( this will give an output as below on the active node)
ACTIVE '/dev/vg-oradata/lv.pdb01' [40.00 GB] inherit
ACTIVE '/dev/vg-oradata/lv.pdb02' [500.00 GB] inherit
ACTIVE '/dev/vg-oradata/lv.pdb03' [500.00 GB] inherit
ACTIVE '/dev/vg-oradata/lv.pdb_arch' [200.00 GB] inherit
ACTIVE '/dev/vg-oradata/lv.dbdump' [49.00 GB] inherit
4. De-Activate all LV's on the active node: * In order to de-activate all the LV’s run the below commands

Syntax: lvchange -an /dev/vg-oradata/lv_name_here (for all LV's)

# lvchange -an /dev/vg-oradata/lv_pdb01
# lvchange -an /dev/vg-oradata/lv_pdb02
# lvchange -an /dev/vg-oradata/lv_pdb03
# lvchange -an /dev/vg-oradata/lv_pdb_arch
# lvchange -an /dev/vg-oradata/lv_dbdump

5. Verifying De-Activated LV’s

# lvscan
inactive '/dev/vg-oradata/lv.pdb01' [40.00 GB] inherit
inactive '/dev/vg-oradata/lv.pdb02' [500.00 GB]
inherit inactive '/dev/vg-oradata/lv.pdb03' [500.00 GB]
inherit inactive '/dev/vg-oradata/lv.pdb_arch' [200.00 GB]
inherit inactive '/dev/vg-oradata/lv.dbdump' [49.00 GB] inherit

On NODE01

6. Activate the LV's on NODE01
Syntax: lvchange -ay /dev/vg-oradata/lv_name_here (for all LV's)
# lvchange -ay /dev/vg-oradata/lv_pdb01
# lvchange -ay /dev/vg-oradata/lv_pdb02
# lvchange -ay /dev/vg-oradata/lv_pdb03
# lvchange -ay /dev/vg-oradata/lv_pdb_arch
# lvchange -ay /dev/vg-oradata/lv_dbdump

7. Mount the File System:
Syntax: mount /dev/vg-oradata/lv_name_here /mount_point_name (for all mounts)
#mount /dev/mapper/vg--oradata-lv.pdb01 /pdb01
#mount /dev/mapper/vg--oradata-lv.pdb02 /pdb02
#mount /dev/mapper/vg--oradata-lv.pdb03 /pdb03
#mount /dev/mapper/vg--oradata-lv.pdb_arch /pdb_arch
#mount /dev/mapper/vg--oradata-lv.dbdump /dbdump/pdb

8. VIP interface activation on NODE01 ( on Node02 we have ifcfg-eth4:0 as the VIP)
# ifup eth4:0
9. Have DBA's start listener and database and verify.
Labels:

Comments

Post a Comment

Popular posts from this blog

How To Manage Dell Servers using OMSA – OpenManage Server Administrator On Linux

OMSA is a web based application to manage DELL PowerEdge Servers. Using OMSA you can perform proactive system monitoring, system diagnosis, troubleshoot hardware issues and configure RAID etc., You can also view and manage hardware’s embedded system management (ESM) log using OMSA. This is an jumpstart guide that explains how to install Dell OMSA on Linux. I have also provided few screenshots of DELL OMSA web application. 1. Download DELL OMSA Go to DELL support website -> click on “Drivers & Downloads” -> choose your server model (in my case, I selected PowerEdge 2850) -> choose the operating system -> scroll-down and expand ‘Systems Management’ -> Click on ‘OpenManage Server Administrator Managed Node’ -> Click on OM_5.5.0_ManNode_A00.tar.gz to download it. You can directly download it to your Linux server as shown below. # mkdir -p /usr/local/omsa # cd /usr/local/omsa # wget http://ftp.us.dell.com/sysman/OM_5.5.0_ManNode_A00.tar.gz 2....
Post a Comment
Read more

What is Cybersecurity Risk? Definition & Factors to Consider

  Cybersecurity risk has become a leading priority for organizations as they embrace digital transformation and leverage advanced technology solutions to drive business growth and optimize efficiencies. Additionally, many organizations are increasingly reliant on third-party and   fourth-party vendors   or programs.  In this post, we’ll explore what cybersecurity risk is and take a look at some key cybersecurity risk factors that organizations across all industries should keep in mind as they build and refine their   cybersecurity risk management strategy .   What is cybersecurity risk? Cybersecurity risk refers to   potential threats and vulnerabilities   in digital systems. It encompasses the likelihood of a cyberattack compromising data or systems, leading to financial,   reputational , or operational damage. A few examples of cybersecurity risks include   ransomware ,   malware ,   insider threats ,   phishing attacks ...
Post a Comment
Read more

Ephone Hunt Groups and Voice Hunt Groups Comparison

SIP phones support Voice Hunt Groups. SCCP phones support Ephone Hunt Groups, and in Cisco Unified CME 4.3 and later versions, SCCP phones also support Voice Hunt Groups.  Table 69  compares the features of Ephone Hunt Groups and Voice Hunt Groups. Table 69 Feature Comparison of Ephone Hunt Groups and Voice Hunt Groups Feature Ephone Hunt Voice Hunt Group Endpoints Supported SCCP only SIP, SCCP, PSTN, and FXS Parallel Hunt Groups (Call Blast) No (for alternative, see the  "Shared-Line Overlays" section ) Yes Hunt Statistics Support Yes No B-ACD Support Yes No Features such as present-call and login/logout Yes No Thanks & Regards Avinash Pillai URL :  http://avinashpillai.blogspot.com Email: avinashp25[AT]gmail[DOT]com
Post a Comment
Read more