
Security Basics mailing list archives
Re: Windows patch mgmt.
From: Keith Cirelli <kcirelli1 () yahoo com>
Date: Tue, 22 Jun 2004 19:58:00 -0700 (PDT)
Microsoft has white papers and documents on patch management that may be helpful for you. Information at the following should be helpful. http://www.microsoft.com/security/guidance/topics/PatchManagement.mspx I believe the problem and solution may be your approach to patch management. [snip][paste] "the question has come up how do we test all desktops/servers --->after<--- a windows patch has been installed." and "The concern is if the server isn't being used for testing, then we may push a patch to a production server without it being "tested."" Applying patches to any production server without testing it is an "Apply and Pray/Apply at your own risk" approach to patch management. The first thing Microsoft would probably ask you if you called them for support (if something did break), would be, "Did you test it first?" Never push a patch without testing. Period. The one time you get burned for NOT testing and things go south and cost the company money, will be all the advice you need to adjust your approach to patch management, if my advice doesn't. Reacting to a potential issues because something wasn't tested can be much more costly to your time and your company's pocket. Any OS build worth putting into production (especially servers) is worth having a duplicate in a lab for testing against things....such as patches. Not knowing how a patch will effect your builds is dangerous for an admin these days. One can err to the side of "limited resources" as to why testing may not be performed PRIOR to installing a patch, but try to imagine what happens when that one patch...that conflicts with something, wasn't tested prior to installation, gets installed and blows up a standard service on a critical machine/server and production is effected. That is not a fun scenario. Someone has to be responsible if production is affected...correct? Who made the decision to apply the patch? Why? Who tested it? See where I'm going? CYA is a reason to test, your own sanity is another, operational uptime might be another? Dedicating the time up front to apply a patch in a lab environment that, as exactly as possible simulates ALL workstation and server builds is an essential part of the Patch Management process. Good workstations can be built as servers just fine for testing purposes in a lab. Merely getting the patch installed to 350 machines is only a part of the process. Determining the validity of a patch in your environment and thoroughly testing that patch BEFORE you deploy it is just as important. It's essential to simulate all of the production builds you have in a lab, you do have standards for builds, riiiiiight?)...make these lab builds as close as possible to production builds. That way you will know if a patch that you have validated as necessary for your environment, works or blows things up BEFORE you blindly apply it. OS services can break, applications can break, network functions can break...when patch's go bad. If a patch does cause adverse effects in your testing, Microsoft is pretty good about providing work arounds that can be very helpful to closing the security hole a poorly written patch is designed to fix until they fix it themselves. Here's what is suggested for time-frames before applying patches... Severity rating Recommended patching time frame Recommended maximum patching time frame Critical Within 24 hours Within two weeks Important Within one month Within two months Moderate --Recommended patching time frame Depending on expected availability, wait for next service pack or patch rollup that includes the patch or deploy the patch within 4 months. --Recommended maximum patching time frame Deploy the patch within six months Low --Recommended patching time frame Depending on expected availability, wait for next service pack or patch rollup that includes the patch, or deploy the patch within 1 year --Recommended maximum patching time frame Deploy the patch within one year, or choose not to deploy at all Hope that helps Keith --- bob martin <bobmartin_613 () hotmail com> wrote:
Hello all. Basic patching question for you. We have a small environment (approx. 300 desktops and 50 servers) and the question has come up how do we test all desktops/servers after a windows patch has been installed. Given that the networking/desktop team consists of 6 people, I'm a bit stumped on how we can do this efficiently. We use St. Benard's Update Expert to push out the patches and to verify they've been installed. Currently we push to a QA environment and let it soak for a week or two while it's being used for it's normal functions. The concern is if the server isn't being used for testing, then we may push a patch to a production server without it being "tested." Any suggestions would be very welcomed. Any more, there's so many windows patches that it's almost a full time job for one person to manage them. Thanks. Bob _________________________________________________________________ Is your PC infected? Get a FREE online computer virus scan from McAfee® Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 --------------------------------------------------------------------------- Ethical Hacking at the InfoSec Institute. Mention this ad and get $545 off any course! All of our class sizes are guaranteed to be 10 students or less to facilitate one-on-one interaction with one of our expert instructors. Attend a course taught by an expert instructor with years of in-the-field pen testing experience in our state of the art hacking lab. Master the skills of an Ethical Hacker to better assess the security of your organization. Visit us at: http://www.infosecinstitute.com/courses/ethical_hacking_training.html ----------------------------------------------------------------------------
--------------------------------------------------------------------------- Ethical Hacking at the InfoSec Institute. Mention this ad and get $545 off any course! All of our class sizes are guaranteed to be 10 students or less to facilitate one-on-one interaction with one of our expert instructors. Attend a course taught by an expert instructor with years of in-the-field pen testing experience in our state of the art hacking lab. Master the skills of an Ethical Hacker to better assess the security of your organization. Visit us at: http://www.infosecinstitute.com/courses/ethical_hacking_training.html ----------------------------------------------------------------------------
Current thread:
- Windows patch mgmt. bob martin (Jun 21)
- Re: Windows patch mgmt. steve (Jun 22)
- Re: Windows patch mgmt. Joe Polk (Jun 23)
- Test Lab Help sEc nErD (Jun 24)
- Re: Windows patch mgmt. Joe Polk (Jun 23)
- Re: Windows patch mgmt. Keith Cirelli (Jun 23)
- <Possible follow-ups>
- RE: Windows patch mgmt. Britton, Jeff B. (Jun 21)
- RE: Windows patch mgmt. Depp, Dennis M. (Jun 22)
- Re: Windows patch mgmt. Murad Talukdar (Jun 23)
- Re: Windows patch mgmt. pingywon MCSE (Jun 23)
- RE: Windows patch mgmt. Paul Ryan (Jun 24)
- RE: Windows patch mgmt. Kymer, Daniel (Jun 23)
- RE: Windows patch mgmt. Depp, Dennis M. (Jun 23)
- Re: RE: Windows patch mgmt. Warren V Camp (Jun 23)
- RE: Windows patch mgmt. Depp, Dennis M. (Jun 23)
- Re: Windows patch mgmt. Ansgar -59cobalt- Wiechers (Jun 25)
(Thread continues...)
- Re: Windows patch mgmt. steve (Jun 22)