Primary SQL Server is inaccessible

Description

This alert is triggered when the primary SQL Server is inaccessible, indicating a potential outage or connectivity issue that would severely impact Relativity functionality.

Resolution Guidance

Relativity becomes unavailable — users cannot log in or access any workspaces.
Document review, audits, and workspace configurations are inaccessible.
All background jobs (indexing, imaging, processing, production) fail to start or remain pending.
Agents and services stop functioning as they rely on SQL for task management.

How To Resolve

There can be multiple reasons why SQL Server becomes inaccessible. While there is no single guaranteed fix, you can try the following steps to investigate and resolve the issue:

Log into Kibana using valid credentials.
If access is unavailable, contact a Relativity administrator within your organization.
Upon successful login, Navigate to 'SQL Overview' dashboard.
On the dashboard, review the Windows Services Health Check section to identify the SQL Server if the services are not running.
Check SQL Server status- Log into the SQL Server and ensure the SQL Server (MSSQLSERVER) service is running.
Verify connectivity - Ensure that the SQL Server is reachable from the Relativity Web, Agent, and Worker servers (e.g., using ping or telnet on port 1433)
Review logs - Inspect SQL Server logs and Windows Event Viewer for any critical errors or root causes (e.g., disk space, authentication, service crashes).
Verify if there have been any updates to SQL Server user credentials and ensure the EDDSDBO user is correctly configured.
Engage a Relativity database administrator from your organization - Involve the database administration team for further investigation and resolution.

Alert Details

Alert Condition Details

Name	Value	Description
Rule Type	Elastic Query
Data View	metrics-*
Filter Query	relsvr.sqlserver.running : 0	Primary SQL Server is inaccessible
Group	Count	number of Primary SQL Server is inaccessible documents
Threshold	> 0	Count greter than 0, alert triggers
Time Window	90 sec	Verified data for last 90 sec
Frequency	30 sec	Checks for each 30 seconds

Alert Metric Details

Metric Name: relsvr.sqlserver.running

Metric Description: Alert triggers on primary SQL Server is inaccessible for last 90 seconds.

Metric Attributes:

Attribute Name	Description	Value
labels.name	Name of Database	EDDS
labels.state		accessible/inaccessible