Question

Snow customers would like to understand what is required when preparing for a DR scenario for Snow Spend Optimizer products, this includes:

• Snow License Manager (SLM)
• Snow Inventory Server (INV)
• Snow Integration Manager (SIM)

Answer

Before investigating Snow in a DR scenario, it is important that all of your business stakeholders are informed of this planning activity (including the Disaster Response Team). Customers should have internal meetings with their stakeholders to understand the following:

• Current DR Strategy
• Recovery Point Objective (RPO)
• Recovery Time Objective (RTO)
• What is the current Complexity of the infrastructure set-up?
• What's is their current solution for backup/restore?
• What is the minimum backup retention period in place?
• For other applications in the Data Center, what processes are being followed?
• Have DR Test plans been created?
• Has the customer a risk management matrix created?
• What is the communication path to users during DR (outage) time?
• Are existing DR procedure/protocol created already?
• Has the customer the ability to create a test environment to replicate the scenario?
• What components are in between (i.e Loadbalancers)?

We recommend that all customers follow their own internal DR strategy & plan in conjunction with their Business Continuity Management Plan. Snow products do not currently support High Availability (applies to SLM 8, SLM 9 and INV 5) & should be removed from SQL Always On as it is not officially supported.  For other versions not listed please check the Snow System Requirements for further advice.

We recommend that as part of the database maintenance & good practice that daily incremental backups are taken with a weekly full backups for all Snow databases. These jobs should be added into the SQL monitoring tool & checked to ensure they have ran successfully each day.

As part of a customers overall DR planning & strategy, they must ensure that sufficient test plans are created to test core functionality in the product. These should be co-ordinated with the company's Snow administrator as they will the best knowledge as to what functionality to test within the platform. Some example use cases to test may be (after fail-over & after fail-back):

• Is the Snow License Manager UI accessible? (Y/N)
• Are agents successfully communicated with the secondary node (ie. Scan & Send works)? (Y/N)
• Are snowpacks successfully processing on the secondary node? (Y/N)
• Can the Snow Administrator log into the Snow Management Access Control Center (SMACC) console? (Y/N)
• Can the Snow Administrator log into the Snow Inventory Administration Console? (Y/N)
• Are dashboards visible within the Console? (Y/N)
• The connectors for the Snow Integration Manager have successfully aggregated? (Y/N)
• All micro-services have successfully started on the SLM server? (Y/N)
• Sub-use cases should be created for each micro-service to ensure functionality - please refer to Best Practice Monitoring Snow Environments document below.
• Are the reports accessible from the Snow UI?
• Do email notifications send successfully?
• Did the Snow Data Update Job run successfully?
• Other core functionality is accessible (please refer to Snow Administrator for more details), this may include:
• Ability to upload entitlements
• Ability to archive devices
• Ability to remove devices
• Ability to create new users or change user permissions.

**The same scenarios should be tested when failing back also**.

As noted previously, DR Scenarios (where possible) should be reviewed on a test platform with DR & mitigation plans in place - following the customers Business Continuity Management plan. The above are guidelines - please refer to your internal teams for further details.

More useful links can be seen below:

https://community.flexera.com/s/article/BestPractice-MonitoringSnowEnvironments
https://docs.snowsoftware.com