|
IT SOLUTIONS
Your full service technology partner!
|
|
|
Check back for more info. Once we learn more, we'll post it here. Q&A:Q. Does this have anything to do with rights? A. Not sure, but I don't think so. Vista limits your rights to the working and private folders so I tried granting the Everyone user and Authenticated Users full rights. No luck. Q. Does this have anything to do with application virtualization? A. I don't think so. I did a quick test where I disabled application virtualization for Paradox and the BDE. No luck. I also tried installing to my second drive (drive D), but no luck regarding the above issue. However, Vista's virtualization has ramifications for Paradox users and I've documented them in the Install Paradox 9 on Vista article. Q. Is this a Corel or Microsoft issue? If it's a Microsoft issue, won't they fix it in a future patch? A. This is CLEARLY a Microsoft issue in my opinion. Yes, Corel probably did something wrong by Vista standards when they developed Paradox 10, but because both Paradox 10 and 11 don't exhibit this issue on previous versions of Windows, what they did was "legal" and Microsoft does indeed need to fix this Vista issue (and many other Vista issues too). I've seen this exact Vista error on other Windows applications that were developed prior to Vista so my opinion is that it's not a matter of "if", it's a matter of "when". However, Corel could and should fix this and the other minor issues too. I have more tests to do and if I figure out a work-around or hear of a work-around, I'll post it here.
|
KB Post: | How to run Paradox for DOS on Vista |
KB Post: | Install Paradox 9 on Windows 7, or even Vista! |
KB Post: | Paradox 9/10/11 Versions (and WPO 2000) |
Comments1 Comments.
Share a thought or comment...
|
First Comment | |
|
Comment 1 of 7
hey mike, i thought local share should always be set to true to help prevent table corruption. Posted 16 years ago
|
|
Comment 2 of 7
Good point. Always a good precaution considering all the corruption issues with Paradox over the years. I haven't seen where Borland acknowledges that Local Share set to False causes corruption in a BDE only system, but it's better to be safe then sorry. Also, Local Share set to True is required for my Workbench for Paradox add-on. One other Paradox developer emailed me about this issue too, so I updated the above article to take a more conservative approach. I changed the Local Share suggested option to be mainly a developer option. Posted 16 years ago
|
|
Comment 3 of 7
Also, today I updated this article with more details on defeating Vista's application virtualization. A real problem when working with the sample files that ship with Paradox and a crucial issue all developers need to know about. Posted 16 years ago
|
|
Comment 4 of 7
From everything that has been written about this exit issue it seems to me that Paradox is attempting to set a Scheduled task during the closing process. This is failing since it is trying to set a v1 task which will not work in the v2 task scheduler. Until Corel patches this process we will be stuck with killing paradox. It would be interesting to see what task paradox is scheduling, unfortunately I am presently stuck with only a Vista box. It would be possible to import the task into the v2 task scheduler but that would probably not fix the issue with paradox waiting for the task to be set via the api. If someone can get an xml export of the task paradox is setting I would be gratefull Posted 16 years ago
|
|
Comment 5 of 7
Dan: You could be right, I don't know. However, I think the issue doesn't have anything to do with what Paradox is doing. My guess is that it has something to do with what Paradox isn't doing, but you could be right. The new Task Scheduler has an automatic "on end of program" event that is triggered whenever an application exits. My guess is that Paradox is not doing something (or is doing something) upon exit that was legal in XP but now causes Vista's Task Scheduler to crash. I haven't had time to look into it yet, but I plan on pursuing this angle by seeing if I can configure Task Scheduler so that the "on end of program" event skips Paradox. If you, or anyone else, has time to dig into Task Scheduler and discovers it. I sure would appreciate a heads up. Posted 16 years ago
|
|
Comment 6 of 7
Q. Does this happen with Paradox Runtime? A. Yes. For example, when exiting Paradox 10 runtime I get the error and becuase I have the Visual Stueio.Net Just-In-Time Debugger installed, I get the following specific exception: An unhandled win32 exception occurred in pdxrwn32.exe [3696]. Posted 16 years ago
|
Latest Comment | |
|
Comment 7 of 7
The main issue with Paradox on Vista on my home machine is the tendency for records with documents (word, rtf, etc) in blob fields to corrupt, causing an error message which does not shut down the program, if one clicks OK. It took quite a while for me to realize that copying database over to either a W7 or XP machine solved the problem. (on those machines only). I read somewhere that the fix released by Corel was to resolve the problem. As you probably know, however, the fix no longer exists. Since I do not want W7 on my home machine, I wonder if there is any way to fix the record corruption issue on Vista. Posted 13 years ago
|
KB Article #101099 Counter |
42655 |
Since 5/20/2008 |
Go ahead! Use Us! | Call: 916-726-5675 | Or visit our new sales site: www.prestwood.com |
ACCOUNT OVERVIEW:
|
CONTACT US:
Office Hours: 8am-5pm | Mon thru Fri
916-726-5675 (office)
916-726-5676 (fax)
916-595-5675 (text msg)
8421 Auburn Blvd, STE 256
Citrus Heights, CA 95610
|
|
©1995-2024 Prestwood IT Solutions.
[Security & Privacy]
|