Emergency SQL Server help

SQL Server down or seriously degraded?

Speak directly to someone who can investigate the database, not a general first-line helpdesk.

Emergencies we handle

When to call

If the business can’t work, or data is at risk, treat it as an emergency.

Database unavailableApplications can’t connect, or a database is offline, suspect or stuck in recovery.
Storage fullA data or log drive has filled and SQL Server has stopped accepting changes (error 9002).
SQL Server won’t startThe service fails after a patch, a restart, a password change or a moved file.
Failed restoreA restore is failing, or a backup you need turns out to be incomplete or damaged.
Severe performance degradationEverything slowed to a crawl at once, often after a deployment, patch or statistics change.
Blocking and deadlocksUsers frozen behind a single session, or transactions repeatedly failing as deadlock victims.
Failed SQL Agent jobsOvernight imports, integrations or backups failing and holding up the day.
Suspected corruptionErrors 823, 824 or 825, or DBCC CHECKDB reporting consistency errors.

What is affected, and when it started
Anything that changed beforehand: patches, deployments, restarts, storage or password changes
The exact error messages, and the SQL Server error log if you can reach it
The SQL Server version, and whether it runs on a physical server, a VM, hosting or Azure
Who can give us remote access, and who can authorise changes

Don’t run DBCC CHECKDB with REPAIR_ALLOW_DATA_LOSS: it can discard data to make a database consistent
Don’t detach a suspect database or delete log files to free space
Don’t shrink files or restart the service repeatedly in the hope it clears
Don’t overwrite existing backups; keep every backup file you have

Each of these can turn a recoverable situation into a much harder one.

What happens next

Talk it through
You describe what is happening to someone who works on SQL Server, and we agree whether and how we can help.
Get access
Your team gives us remote access by whatever secure route you already use.
Stabilise
We get the service back, or protect the data, using the least risky option available – and explain the trade-offs before anything irreversible.
Find the cause
Once the pressure is off, we establish why it happened and what would stop it happening again.
Write it up
A short summary of what happened, what we did and what we recommend.

Not an emergency yet?

Warnings, growing logs and failing jobs are easier to deal with before they become outages. A health check or ongoing support finds them early.

Next step

Call first. If you can’t call, send the details and choose “Emergency issue”.