MS SQL Cluster

Multi-node failover cluster Instance

SQL Server clustering describes a collection of two or more physical servers (nodes), connected via LAN, each of which host a SQL server instance and have the same access to shared storage. Clustering SQL servers provides high availability and protection from disasters when a server hosting the SQL Server instance fails.

Always on failover cluster instances

SQL Server Always On Failover Cluster Instances (FCIs) use Windows Server Failover Clustering (WSFC) to provide high availability at the server instance level. An FCI is a single instance of SQL Server that is installed across WSFC nodes to provide high availability for the entire installation of SQL Server.

Always On availability group

Always On availability groups feature a high availability and disaster recovery solution that provides an alternative to database mirroring, and maximizes the availability of a set of user databases for an enterprise. This supports a failover environment for a discrete set of user databases, known as availability databases.

The availability group supports a set of read-write primary databases and one to eight sets of corresponding secondary databases. Optionally, secondary databases can be made available for read-only access and/or some backup operations. Availability group fails over at the level of an availability replica. Failovers are not caused by database issues such as a database becoming suspect due to a loss of a data file, deletion of a database, or corruption of a transaction log.

Pre-requisites

• The application would require classic gateway 12.0 and above.

• There is no version dependency for cluster gateway.

• Powershell cmdlets have the following prerequisites:

  • Windows domain User should be able to do powershell remoting
    Enable-PSRemoting -Force
  • Windows domain user should be added to “Remote Management users” group net localgroup “Remote Management Users” /add < user >
  • Windows domain user should be added to “Performance monitor users” group net localgroup “Performance monitor users” /add < user >

• Granting Remote DCOM Rights

  • To grant users DCOM rights, log on to each monitored system and complete the following procedure:
    • Go to command prompt and enter dcomcnfg
    • Navigate to component services > computers > My computer and then right click and select Properties. Then go to the COM Security tab.
    • Under Access Permissions, go to edit limits and add the domain non-admin user and enable both local and remote access then click OK.
    • In Launch and Activation permissions, go to edit limits, add the domain non-admin user and check all boxes , and click OK.

• Granting Remote WMI Rights

  • To give the user remote WMI rights, log on to each system to be monitored and complete the following procedure:
    • Go to “computer management” and select “WMI Control” under ‘Services and Applications’.
    • Right-click WMI control and go to properties. Then select the Security tab.
    • Select root and click Security.
    • Add the domain non-admin user, and check the boxes for execute methods, enable account, remote enable, and reas security.
    • Click Advanced, select added non-admin domain user and click edit. Then for “Applies” to select “namespace and sub namespaces” in the dropdown.
    • Click OK 3 times .

• For monitoring services like windows cluster service and other mssql services (sql server, sql agent , sql browser etc.), windows domain users have to be a part of “Local Administrators group” as we are using win32_Service class for fetching the details.

  • refer to below link:
    https://docs.microsoft.com/en-us/powershell/module/microsoft.powershell.management/get-wmiobject?view=powershell-5.1
  • If you do not want to add the user to local administrator group, you can use “Security descriptors” for monitoring the services.
  • For that we have to do the configuration as below:
    Refer to 3. Enable Windows Service Monitoring section of the doc:

• Windows domain user should be granted read-only access to cluster
  Grant-ClusterAccess -User <domain\user> -Readonly

SQL authorization permissions:

• For monitoring some metrics, we are using JDBC. For JDBC connections we are supporting mixed authentication ( both NLTM and SQL server authentication)
• The minimum privileges for SQL auth are:
  • server role: public - This server role is available by default
  • securables: We need to enable some securables for connecting to SQL instances,connecting to all databases, and also some read-only permissions like below:

 - CONNECT ANY DATABASE
      GRANT CONNECT ANY DATABASE to < user >
 - CONNECT SQL
      GRANT CONNECT SQL to < user >
 - SELECT ALL USER SECURABLES
      GRANT SELECT ALL USER SECURABLES to < user >
 - VIEW ANY DATABASE
      GRANT VIEW ANY DATABASE to < user >
 - VIEW ANY DEFINITION
      GRANT VIEW ANY DEFINITION to < user >
 - VIEW SERVER STATE
      GRANT VIEW SERVER STATE to < user >

Install the integration

1. From All Clients, select a client.
2. Go to Setup > Integrations > Integrations.
3. From Available Integrations, select Adapter >MS SQL Cluster. The Install MS SQL Cluster Integration popup appears.

4. Enter the following information:
   • Name: Name of the integration
   • Upload Logo: (Optional) logo for the integration.
   • GateWay Profiles: Select a gateway management profile to associate with the client.
5. Click Install. The Integration page displays the installed integration.

Configure the integration

1. In CONFIGURATION section, click +Add.
2. On Create Adapter Configuration, enter:
   • Name: Configuration name.
   • IP Address/Host Name: IP address or host name of the target.
   • Notification Alerts: Select TRUE or FALSE.
     Notes:
     • By default FALSE is selected.
     • If you select TRUE, application will handle Critical/Recovery failure alert notifications for Connectivity, Authentication Exceptions.
   • MSSQL Cluster Type: Select AvailabilityGroup or FailoverClusterInstance from the dropdown.
   • SQL Server Authentication Mode: Select WindowsAuthentication or SqlServerAuthentication from the dropdown.
3. From the Windows Cluster Credentials section, select Custom and enter Username and Password.
4. From the SQL Server Credentials section, select Custom and enter Username and Password.
5. From the Resource Types & Metrics section, select the metrics and configure for availability and alert conditions, for Cluster, Server, Group, Storage, & SQL Server.
6. In the Discovery Schedule section, configure how frequently the discovery action should trigger. Select Recurrence Pattern to add one of the following patterns:
   • Minutes
   • Hourly
   • Daily
   • Weekly
   • Monthly
7. In the Monitoring Schedule section, configure how frequently the monitoring action should trigger.
8. Click Save.

After saving the configuration, the resources are discovered and monitoring is done as specified in the configuration profile.

The configuration is saved and displayed on the page.

You can perform the actions manually, like Discovery, Monitoring or Disable the configuration.

Resource information

1. To view the resource information, go to Infrastructure > Resources > Cluster.
2. Use the search function to find your resource and then click on it.

View resource attributes

The discovered resource(s) are displayed under clusters. In this page you will get the basic information about the resources such as: Resource Type, Native Resource Type, Resource Name, IP Address etc.

Note: The children cluster nodes are displayed under Components, click any children node and navigate to Metrics to view the individual metrics to the children node.

View resource metrics

To confirm MS SQL Cluster monitoring, review the following:

• Metric graphs: A graph is plotted for each metric that is enabled in the configuration.
• Alerts: Alerts are generated for metrics that are configured as defined for integration.

Supported metrics

Supported product versions

• The application is validated on Windows Server 2019 and Microsoft SQL Server 2019 (RTM) - 15.0.2000.5 (X64) for both MSSQL failover instance & availability group clusters.

Risks, Limitations & Assumptions

• This sdk application only supports mssql-cluster which is built upon windows failover cluster.

• If an application is uninstalled or application configurations are removed, then the resources associated with that particular application/application configuration will also get deleted along with agent installed devices.

• You have to provide credentials in “SQL Server Credentials” section even though you opted for “WindowsAuthentication” as authentication mode while adding configuration.
• For Cluster object discovery and monitoring implementation, we are considering the object which has “Name” equals to Cluster Name in Get-ClusterResource response
• For ClusterGroup monitoring implementation, we are considering the object which has Name as Cluster Group in Get-ClusterGroup response All the failover status metrics possible instance values are 0 - if there is no change in OwnerNode, 1 - If there is a change in OwnerNode, 2 - If no OwnerNode
• Application can handle Critical/Recovery failure alert notifications for below two cases of when you enable Notification Alerts in configuration:
  • Connectivity Exception
  • Authentication Exception
• If you enable agent monitoring templates on the Cluster/Node resource, you might see the duplicate metrics with different naming conventions.
• While trying to fetch the node ip address we receive multiple node ips, which include many local ips and actual ips (example: lets assume actual node ip is 10.1.1.1 when trying to fetch the details we will receive two ips one associated with custer(192.168.0.0) and other is the actual ip). to identify the actual node ip address from the list of ip addresses received we assume that node ip address is part of the same subnet of cluster ip address, which means if cluster ip is 10.1.1.1 then node ips will be 10.1.X.X.
• We have provided the provision to give Cluster Ip Address OR HostName in configuration, but hostName provision will work only if the host name resolution works.
• Component level threshold configuration is not possible
• Resource level metric threshold customization and frequency setting are not possible
• Usability issues in application configuration page while adding/editing
• Optional configuration parameters cannot be defined
• Application upgrade is manual process without version change