fbpx
CUSTOMER TIPS

Server Reinstallation

Description

A server, such as Storage Server or Connector, may need to be reinstalled for a CentOS upgrade from version 6 to version 7, or from a different operating system. This procedure requires good system skills. Otherwise reach Scality for a service request.

Procedure

Follow these steps to reinstall a server:

  • Make a Backup
  • Re-Install Operating System
  • Prepare Server with requirements
  • Install Scality packages
  • Verify

The plan will work for every component (supervisor, node or connector). The procedure has to be applied server by server.

Make a backup

Use the Scality Backup Tool (scality-backup). This tool will backup most of the common scality configuration files into /var/lib/scality/backup.

The following system settings need to also be backed up:

  • /root & /home (check the root home directory if there is data to backup and other users as well)
  • /etc
  • /opt
  • /var/lib/elasticsearch
  • A list of all installed packages

For the supervisor:

  • /srv

In a RING version 6 environment, some modifications may not be stored into the defined Scality pillar. Make sure we have the correct list of the roles of each servers:

$> salt ‘*’ grains.get roles > /var/lib/scality/backup/salt_grains

Note

Make sure that the /etc/fstab file is correctly configured according to the UUID‘s of disks drives, patition, etc and not according to the labels of these devices.

example

(old) $> scality-backup
(old) $> cp -r ~/ /var/lib/scality/backup/root
(old) $> cp -r /etc /var/lib/scality/backup/
(old) $> rpm -qa > /var/lib/scality/backup/all_installed_packages
(old) $> tar cvfz /tmp/scality_backup.tar.gz /var/lib/scality

Re-Install The Operating System

Before the operation, review Scality requirements about each component.

  • When the supervisor is managed, there will be no monitoring.

  • When the node server is managed, the nodes need to leave before the re-installation, see Performing Server Maintenance.

  • When managing a connector, ensure there is not traffic on it or it may cause an impact on the service during the maintenance.

Make sure that requirements are applied such as the /var partition over 200GB. No IP address will change.

Prepare Server With Requirements

Review all Requirements and Recommendations

  • Disable the firewall

  • Disable SELinux

  • Enable one-shot IRQbalance

  • Install tuned with a latency-performance profile

  • Disable NUMA and THPs

  • sysctl configurations

  • Make sure NTP is enabled and synchronised

Reboot the server after applying those modifications.

Install Scality Packages

1 Copy the backup-ed archive to the server and extract it.

$> tar xvfz backup_.tar.gz -C /

All previous backup-ed files will be restored into /var/lib/scality/backup.

2 Install salt-minion

$> yum install salt-minion

3 Restore the salt configuration:

$> cp -rf /var/lib/scality/backup/etc/salt/* /etc/salt

4 Ensure sure salt-minion is enabled and restart the service:

$> systemctl enable salt-minion
$> systemctl restart salt-minion

5 Through the supervisor, ensure it is reachable:

(sup) $> salt '' test.ping
(sup) $> salt '' saltutil.sync_all refresh=True

6 Apply scality credentials

$> salt-call state.sls scality.credentials
$> salt-call saltutil.sync_all refresh=True

7 Apply repositories and requirements

$> salt-call state.sls scality.repo
$> salt-call state.sls scality.req

8 Ensure the Scality repositories are available

$> yum repolist | grep scality
!scality-internal Scality HTTP repository (internal) - RHEL 7 - 179
!scality-offline Scality HTTP repository (offline) - RHEL 7 - 875
!scality-saltstack/7/x86_64 Scality Saltstack repository 101

9 Install the Scality packages

## Review the list of previous scality packages
$> egrep 'scality|elasticsearch|java|httpd' /var/lib/scality/backup/all_installed_packages
## From the list, installing the needed packages
## Example:
$> yum install scality-sproxyd scality-srebuildd scality-node elasticsearch httpd java

Ensure all Scality services are stopped before the restoration.

10 Restore configuration files

## Ignoring backup-ed fstab file on scality-biziod
$> mv /var/lib/scality/backup/scality-biziod/etc/fstab /var/lib/scality/backup/scality-biziod/etc/fstab.before_reinstall
## Restoring all configuration files
$> cp -rf /var/lib/scality/backup/scality-*/* /
## Restoring elasticsearch configuration
$> cp -rf /var/lib/scality/backup/etc/elasticsearch /etc

11 Restore fstab and mountpoints

## Restoging fstab
$> grep scality /var/lib/scality/backup/etc/fstab >> /etc/fstab
## Creating all mountpoints
$> for mountpoint in $(grep scality /etc/fstab | awk '{print $2}'); do mkdir -p $mountpoint; done
## Mouting all mountpoints
$> for mountpoint in $(grep scality /etc/fstab | awk '{print $2}'); do mount $mountpoint; done
## Check:
$> mount | grep scality
$> lsblk

12 Enable all services

$> systemctl enable scality-sproxyd
$> systemctl enable scality-srebuildd
$> systemctl enable httpd
$> systemctl enable elasticsearch
$> systemctl enable scality-sagentd
$> systemctl enable scality-node
$> systemctl start scality-sproxyd
$> systemctl start scality-srebuildd
$> systemctl start httpd
$> systemctl start elasticsearch
$> systemctl start scality-sagentd
$> systemctl start scality-node

Note: From the list of installed packages previously got (only example here).

13 Re-Install S3
Only if the server had S3 role. Through Federation reinstall it by limiting with the server.

(Federation) $> ./ansible-playbook -i env/{{EnvDir}}/inventory run.yml --limit mdX-clusterY

Verify

Ensure the supervisor reports correctly the server once services are started. – Review the performance – The nodes should be online in the supervisor GUI, with all disks OK (ALL RINGS) – No node should have the ARC_REBUILD_NOK status (ALL RINGS) – The services sagentd, sproxyd and srebuildd should be running with no error – The number of Missing Objects should not increase in the supervisor GUI (ALL RINGS) – The elasticsearch cluster should be complete:

(sup) $> curl http://localhost/api/v0.1/es_proxy/_cluster/health?pretty

If this does not solve your issues please get in touch with support

SAN FRANCISCO, USA

149 New Montgomery Street, Suite 607
San Francisco, CA, 94105
Email: sales.us@scality.com
Telephone: +1 (650) 356-8500
Fax: +1 (650) 356-8501
Toll Free: +1 (855) 722-5489


PARIS, FRANCE

11 rue Tronchet
75008 Paris, France
Email: sales.eu@scality.com
Telephone: +33 1 78 09 82 70


WASHINGTON, D.C., USA

43777 Central Station Drive, Suite 410
Ashburn, VA 20147, USA
Email: sales.us@scality.com
Toll Free: +1 (855) 722-5489

TOKYO, JAPAN

Otemachi Bldg. 4F, 1-6-1, Otemachi
Chiyoda-ku Tokyo, 100-0004 Japan
Email: sales.japan@scality.com
Telephone: +81-3-4405-5400


LONDON, UNITED KINGDOM

Quadrant House, Floor 6
4 Thomas More Square
London, United Kingdom, E1W 1YW
Email: sales.eu@scality.com
Telephone: +44 203 795 2434

Products

The most powerful data storage platform.

Protect, search and manage your data on any cloud.

RING

ON PREMISES, PRIVATE CLOUD STORAGE SOLUTION

ZENKO

DATA MANAGEMENT ACROSS
MULTIPLE CLOUDS

NAS ARCHIVER

TIER INACTIVE DATA FROM NAS TO SCALITY RING

Hardware Alliances

Delivering fully integrated solutions.

Learn More

Read about Scality data storage and management solutions.