If this is a Windows XP SP #2 (or, for that matter, Windows Server 2003 SP #1 RC #1) issue?
I have my own set of programs having problems with it!
(Oddly only some (2 of the total 30 I put out publicly), which is rather odd because I develop my programs from a single generic template here using Delphi.
I suspect it's a .VCL file I use in the 2 that break down, but DO definitely have it "zeroed into" this new AMD CPU opcode against "buffer overflow" exploits. This I know now for a fact from traces of my apps via plain-old DrWtsn32.exe dumps of the abends/errors!
It may be the fact that I use GIGANTIC stack-sizes on the 2 programs of mine that have this issue as well. Ordinarily, that translates out to GIANT paged-out virtual memory use, but should promote program stability...
It's not doing added stability @ ALL though the second I set Windows Server 2003 SP #1 RC #1 to protect ALL programs, not just Microsoft OS files! )
I first suspected the fact I use "exe compressors" on my code for faster load time from disk as well as protection of the apps against disassemblers (this helps here) & to aid in built-into my apps rudimentary antivirus protection (I check size of app on disk on loads usually in MOST of my work - if diff. than hardcoded value of size? App is considered infected... & shuts down etc. )
I think I was wrong about that - yes, it happens... anyhow, read on:
So, that all said? What would I do if you're having XP SP #2 hassles??
Go here in Start Menu-> Control Panel -> System Icon-> Advanced Tab -> Performance Group/Settings Button -> (next screen that pops up) Data Execution Prevention Tab & check off the "Turn on DEP for essential Windows Programs and Services ONLY" (vs. the other option of "Turn on DEP for all programs and services except those I select")
This works for me here just fine on my own work, & others from people like System Internals (whose work may be experiencing hassles here due to the nature of their construction, which many of their progs use user mode device drivers &/or kernel mode device driver work... Windows Server 2003 has a NEW service called "Windows User Mode Driver Framework" I suspect in their work's case, is in their face... but, that's their work, not mine & I can only speculate on why some of their stuff is backfiring here).
Try that, sorry for the long-winded explain, but should help!
(Has for me)
* Let us know how this goes for you XP SP #2 users & Windows Server 2003 SP#1 RC #1 users too amongst you all...
APK
P.S.=> It's NEW technology for MS, so we have to cut them some slack... the idea is good, but already folks are finding holes in it, bugs, & even how to exploit it as of 2-3 days ago on SlashDot by some Russian guys iirc here:
http://www.maxpatrol.com/ptmshorp.asp
& yet another .NET hole surfaces (this is one of many so far):
http://www.zdnet.com.au/news/security/0,2000061744,39179932,00.htm
(Some of the finest "hacker/*****ers" there are imo, are from the eastern block, so I trust what they say quite alot in regard to exploits & such personally. Ms will try to cover for it, I don't blame them because sometimes the circumstances required for some of these are pretty "outrageous" from what I recall over time on these & other 'holes' but, again they're new technologies - expect some "perfection time" for them to be 110% solid for the MOST part)... apk
2005-01-19 22:27:33
2005-01-20 12:05:13
2005-02-05 00:19:08
View profile
Mark this post as solution (topic starter only)
Reply with quote to post
Edit post
Delete post (admin only)