I'm new to the IT industry and I'm definitely not a sysadmin (I'm a road construction technician by trade). So please correct me if I mix up terminology, sysadmin concepts, etc.

My organization's IT department is very small. So small, that we seem to be barely treading water when it comes to keeping our IT systems "up".

The thing that seems to cause us the most problems is Windows Server updates. I'm told that the updates aren't totally automatic (they don't install the minute Microsoft releases them), but they are scheduled to happen automatically on a monthly basis -- on weekdays after-hours.

We come across issues like the following:

  1. The virtual application server that contains the WebSphere JVMs for our work order management system was updated last night (via the 1-month schedule).

  2. Today, when users started using the work order management system, we got a bunch of integration/java errors from the JVM.

    • This issue was clearly caused by the updates. We've never had that issue before, and it happened right after the updates were applied.
  3. Our sysadmins restarted the server which seemed to solve the issue. We dealt with the failed integration messages, etc. and life goes on...at least until the next update happens.

  4. That's just one example. We actually had a handful of systems that had similar issues after the update last night.

I know that my organization isn't the only one that has problems with windows updates. It seems to be pretty widespread. But my question is:

Is there some sort of basic technique for handling updates that my organization might have overlooked?

For example, it occurred to me that we could:

  1. Manually apply the updates on Saturday mornings
  2. Test the crap out of ALL our systems
  3. Have ALL weekend to deal with issues and restart servers if we need to (instead of doing it live during business hours).
  4. We could also consider restarting ALL servers each weekend, and maybe even do updates at that time, so that fewer updates were applied each time, which might be less risky.

The major drawback to the ideas above is manpower (money). We don't have it. It would take a significant shift in organizational thinking & budgeting to do those things, so I just wanted to ask: what's the right way to handle these updates? Could the above ideas solve the majority of our issues?

New contributor
User1973 is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
  • The "right way" might well have very little to do with the windows OS and very much with improving the general quality of the business-critical software that runs on top. You see, software targeting the windows OS should not fail on things that are normal during operation of the windows OS. From the outside it is hard to tell though, experienced admins tend to be more careful about "it happened right after X, so it must be caused by X" conclusions. – anx 2 days ago

Sorry to hear about all the errors in your production environment due to the Windows server updates. I dont work with windows servers but almost pretty sure those updates can be disabled to do not apply automatically. What normally happens is that there are two environments one for testing and one for production so you dont deploy anything that has not been tested first in your test enviroment.

You may also find interesting topics such as DevOps ITIL Cloud Computing.

If the information was helpful please don't forget to up vote or accept the answer.

New contributor
rriovall is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.

Your Answer

User1973 is a new contributor. Be nice, and check out our Code of Conduct.

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.