Free Webinar
October 11 | 11am PT / 2pm ET
Learn how to build an IT infrastructure of your dream
with Dell EMC PowerEdge 14G servers
Speaker: Ivan Talaichuk, Pre-Sales Engineer, StarWind

StarWind Virtual SAN Maintenance Mode

Published: January 17, 2018

INTRODUCTION

This technical paper describes how to configure Maintenance mode for HA devices using StarWind Management Console. In any production environment, there are situations that require an administrator to stop production services for a while to perform certain operations on production nodes. This may be a planned nodes servicing, their planned stops, or planned power outage, etc. After turning on the nodes and resuming their production operation, it is nice to have fully-synchronized data on all nodes in order to avoid long-lasting synchronization process that can take a considerable amount of time in most production environments.

Enabling Maintenance mode in StarWind Management Console before planned shutdowns saves time required for data synchronization between nodes after exiting Maintenance mode and simplifies the process of making any changes in highly- available production environments.

To simplify the production startup after the scheduled service interruption, just turn on the Maintenance mode before performing various actions (shutdown, reboot, etc.) on the nodes. Even if nodes are turned off incorrectly after entering the Maintenance mode, all data will be safe. After finishing the operations on the nodes, just exit the Maintenance mode. As soon as this has been done, production can be launched without any need to wait for either fast or full synchronization. The only thing that should be checked is connectivity between the devices under maintenance.

When a device enters the Maintenance mode, it should contain fully-synchronized data on all HA nodes. Maintenance mode for the selected device will be enabled/disabled on all nodes simultaneously. When an HA device is in Maintenance mode, connection to this HA device is blocked on any node containing it to ensure integrity of data till the Maintenance mode is off. As a result, the synchronization before starting the HA operation is avoided, as data on all nodes is completely identical to what had been there before entering the Maintenance mode.

The full set of up-to-date technical documentation can always be found here, or by pressing the Help button in the StarWind Management Console.

For any technical inquiries, please, visit our online community, Frequently Asked Questions page, or use the support form to contact our technical support department

Maintenance mode configuration steps

NOTE: Only HA device you select will enter into Maintenance mode. In case you need to reboot or shut down the nodes, Maintenance mode must be enabled on all HA devices the node has. Make sure that you have shut down production VMs that use the storage before devices enter the Maintenance mode. Maintenance mode also should be enabled in a hypervisor environment. The StarWind Maintenance mode for the selected device will be enabled/disabled on all nodes simultaneously.

1. Before entering StarWind device into the Maintenance mode, it must be in the Synchronized state on all nodes. For this purpose, open StarWind Management Console, connect to the servers and check the device state by clicking on it.

StarWind Virtual SAN Maintenance Mode

2. Select “Enter Maintenance mode” by clicking the StarWind device.

StarWind Virtual SAN Maintenance Mode

The device will be set into Maintenance mode on all StarWind nodes.

StarWind Virtual SAN Maintenance Mode

3. After all planned servicing actions are done, select “Exit Maintenance mode” by clicking the StarWind device.

StarWind Virtual SAN Maintenance Mode

Exiting the Maintenance mode allows the nodes to start working in the normal operation mode without either full or fast synchronization.

CONCLUSION

StarWind Maintenance mode is a StarWind Virtual SAN feature allowing to avoid full synchronization process after planned nodes shutdowns for servicing, planned power outage etc. It saves the precious time and efforts on production startup as well as averts possible performance drops during the full synchronization process after the scheduled service interruption. Being enabled on all devices, it ensures stored data integrity, even if their hosts are shut down incorrectly.