General discussion

Locked

Windows OS hotfixes/patches and Databases

By MirrorMirror ·
I am looking for advice. I am the SQL Server DBA at our company. Our network admins want to apply Wondows 2000 Server and Windows 2000 Advanced Server hotfixes and other service packs to servers that are running SQL Server 2000. I am looking for a way to confirm that the updates will not affect SQL Server. The patches have been applied to a development server(without my knowledge) and SQL Server is running fine. But, I would like to be able to research these patches first, test them on a server running SQL Server and then approve them to move to a production server. Is there anywhere that lists known hotfix issues with SQL Server? I have been reduced to checking the KB articles on TechNet and looking for caveats on each to see if there are any issues with SQL Server.

Any advice is welcome.

This conversation is currently closed to new comments.

5 total posts (Page 1 of 1)  
| Thread display: Collapse - | Expand +

All Comments

Collapse -

My Experience

by jdmercha In reply to Windows OS hotfixes/patch ...

I've got a 2000 server running SQL server and I have it set to auto update. It's been set this way for about two years now. I've not had any problems. But I have not gone to that latest service pack. (I don't have the disk space.)

Collapse -

Software testing

by JamesRL In reply to Windows OS hotfixes/patch ...

The only way is to come up with a set of tests. Hopefully when someone developed the application in the first place, they created a series of tests - functional tests, system tests, communications tests etc. At a minimum, you would liek the chance to run all of these tests again to see if anything breaks under the new patches - on a test server of course. Better yet run them once before the patches are applied and once after - the software may have changed since the original tests were written. If you have up to date tests for everything, consider yourself lucky.

You might want to go through the KB articles and identify any areas of particular risk - anything areas that were "fixed" by the patches are areas that you want to test more thoroughly.

Never test on a production server, and make sure that your test server is as close as possible to the configuration of the production server after the patches are applied.

James

Collapse -

What your Admins did was right...however

by dafe2 In reply to Windows OS hotfixes/patch ...

IMO you do both. Research & Test. The test is done on either a Dev Server or a Spare Server.

You bring up a good point though --- The constant 'fight' between Developers & Operations.

The best way arround this is to implement Change Management.

That way, the Dev Group and Operations are at the same table on the same page.

Collapse -

Yes they are correct

by MirrorMirror In reply to What your Admins did was ...

I am not trying to fight with our admins about doing OS updates. There really is no struggle going on. As the DBA, I need to know what changes are happening on the servers that run the databases. If an OS update adversely affects the database, I need to be able figure this out before we apply the hotfix/patch to a production server. I agree that to do this in test is best. But, even in test I should be aware of what is going on just in case. I just want to make sure that ports won't be shut down or policy changes put in that will make the databases unavailable.

I am just wondering how other shops handle OS patches/hotfixes on database servers. I have seen some servers running some very finicky stuff that would absolutely stop if .dll's were changed in any way.

I absolutely think that OS updates, especially security updates, should be applied. Change management would be a great thing but so far that has not been implemented in full where I work. As usual, test servers are few and far between. Any advice from all you network admins out there?

How do you communicate OS patches/hotfixes to your DBA's? What is your SOP for applying them? What is the change management process that you use?

Collapse -

Change Management Process (ITIL)

by dafe2 In reply to Yes they are correct

We use ITIL Principles (Google for Pink Elephant or ITIL)

Any infrastructure, server or system changes are routed through a CAB (Change advisory Board) for review by representatives from all business units, workgroups or stakeholders (DBA's).

Concerns or issues are dealt with at that review.

Back to IT Employment Forum
5 total posts (Page 1 of 1)  

Related Discussions

Related Forums