Separate database service, connectivity, authentication, workload and storage issues before changing SQL Server, MySQL, MariaDB or PostgreSQL.
- For connection failures, record the engine, instance, endpoint, authentication method and complete error.
- For blocking or high load, identify the session, transaction and execution context before interrupting work.
- For backup or capacity incidents, examine the backup chain, logs, data files and underlying storage.
Command examples
Adapt placeholders, permissions and target systems before using an example. These examples are documentation; no command runs from this page.
SELECT * FROM sys.dm_exec_requests WHERE blocking_session_id <> 0;SELECT name,state_desc,recovery_model_desc,log_reuse_wait_desc FROM sys.databases;RESTORE VERIFYONLY FROM DISK='D:\Backup\DB.bak';SELECT pid,usename,state,query FROM pg_stat_activity;SELECT pg_size_pretty(pg_database_size(current_database()));Check the database endpoint
Use the actual listener or instance port. TCP 1433, 3306 and 5432 are common defaults for SQL Server, MySQL and PostgreSQL, but instance configuration can differ.
Collect SQL Server evidence
Record service and database state, active requests, blocking and free space. Querying a diagnostic view may require additional monitoring permissions.
Verify and test backups
RESTORE VERIFYONLY checks backup completeness and readability; it does not restore the database or fully verify the data structure. Complete a restore test in an appropriate environment.
Preserve database files
Do not manually remove MDF, LDF, WAL or binary-log files to create free space. Follow the engine’s supported retention and recovery process.
Understand blocking first
Identify the blocking session, its transaction and the impact of rollback before terminating it.
Assess execution evidence
Review the workload and execution plan before adopting a suggested index. An index can change write cost and storage requirements.
Include storage performance
Correlate database waits and response time with storage latency and I/O activity.
Use appropriate privileges
An access error should be diagnosed against the required permission and scope rather than resolved by granting unrestricted administrative access.