Flash9b.ocx crash once again

ok...I tried FF and now instead of getting the error message from IE, I'm now getting one from FF and it crashes....I GIVE UP!!!
Any other options?

quote:

---

Originally posted by: boostm...I dont agree with you that it just happenstance, or 'pure luck' that Flash Player plays nicely with FireFox. And I dont put the whole blame on Adobe...Mozilla and Adobe do not have this problem, and there is absolutely no reason to think that a future version will....

---

A prior post indicates Flash Player isn't always checking return codes for memory allocation calls. They merely assume the call succeeds, then attempt to use the requested memory block. If the allocation call fails (or simply fails to return the expected size block), Flash Player still tries to reference that, which causes an exception and crashes the browser. By definition that is 100% the fault of the caller (flash9d.ocx). This commonly happens in software development. Applications frequently make improper assumptions.

Why would Flash Player not crash on FF? Maybe FF uses a different memory allocation scheme which by chance succeeds on the allocation calls made by flash9d.ocx. Maybe that same scheme has other downsides such as slower performance, more memory fragmentation; it just doesn't crash (yet).

The bottom line is the caller (Flash Player in this case) must always check for success/fail and not make non-documented assumptions about the behavior.

This is a very common scenario in software development. Failing to check for success/fail on an allocation call often causes random crashes of exactly the type seen here.

It is very easily possible that updates to FF could revise the memory management scheme, which would cause Flash Player to crash it just like IE7.

That's why Adobe must still fix the underlying problem, and why using FF isn't guaranteed to avoid this problem in the future.

.

quote:

---

Originally posted by: boostm if I open a new tap, or refresh an existing tab, and rest my mouse on the scrollbar after scrolling to a portion of the screen, the focus of the mouse becomes 'sticky', and it behaves as though the mouse left button is pressed when it isnt..

---

I've never experienced that. My first suggestion is download the latest mouse drivers for the type of mouse you have. Then check in control panel -> mouse, and verify the properties, button assignments, are as you think.

>>So don't be hesitant to try FF, Flash Player doesn't currently seem to crash that browser like it does IE7. However this is likely due to pure luck<<

I dont agree with you that it just happenstance, or 'pure luck' that Flash Player plays nicely with FireFox. And I dont put the whole blame on Adobe.. Clearly there is something going on between Microsoft and Adobe that is preventing them from getting their 'signals straight'.. Whether its a lack of sharing of key code modules or something different.. But the fact is, Mozilla and Adobe do not have this problem, and there is absolutely no reason to think that a future version will. While Adobe shares in some of the blame, they dont share in all of it.. That Mozilla has an internet reader which works so nicely with Adobe is proof in and of itself.

But, Joema, since youre using Firefox, maybe you can answer a related question. Im finding some anomalous behavior with regard to scrolling.. Im finding that if I open a new tap, or refresh an existing tab, and rest my mouse on the scrollbar after scrolling to a portion of the screen, the focus of the mouse becomes 'sticky', and it behaves as though the mouse left button is pressed when it isnt.. The result is, simply moving the mouse with no buttons depressed causes the tab contents to scroll just as if I were dragging the scroll bar . It can be quite annoying.. To make it stop, I have to explicitly click the left mouse button somewhere in the open window to make it 'lose focus' and make the scrollbar become ' un sticky'... Have you ever seen this behavior? It happens quite regularly, and no adjustment to any of settings will change it.. It Never happened with IE7. Weird.. Too bad, too, cause its the Only thing I dont like about Firefox at the moment. Ive read some of the development errata, and there is reference to something that sounds like this, but not a solution.

I posted this before, but wanted to post it again to make sure you guys realize you have an option.. If Adobe wont fix this problem in a timely manner, and you find yourself having to deal with more crashes than youd care to, you DO have a way out.. Simply download and install Mozilla Firefox, and start using it instead of IE. I did that almost a month ago. The look and feel of Firefox is Very similiar to the tab interface of IE. And, ive not had a single crash, no matter how many simultaneous tabs Ive got open.

Its good to know that we DO have options.. We dont have to wait for either Microsoft or Adobe to do something about it. Since we do have such a good alternative, in the end, its our own fault if we continue to put up with the rotten support weve been getting for this problem.

And, the nice thing is, we dont have to uninstall IE or make any changes therein at all. All you need to do is install Firefox, and use it! Temporarily (if you want) make it your primary internet reader, and use it until youre convinced it really is the better way. As a security blanket, IE is still sitting there in its closed, iconic state, ready to jump back in (and crash your pc :) ) whenever you want it back.. But my bet is, you never will!!!

Senior Adobe support or development engineers need to use the existing array of debugging tools to fix this problem.

An earlier post indicated Flash Player wasn't properly checking return codes when it allocated memory, and always assumed success, even if the allocation call failed. That typically causes intermittent, random failures of the type seen here. Not checking return codes is poor programming practice, and obviously the first step is fixing all those.

The next step is using a run-time analyzer such as BoundsChecker or Purify. That can be done in-house, and will catch other memory errors. I don't know if that's possible for an ocx product, but it should be considered.

While it's nice to have a perfectly reproducible scenario, sometimes this isn't possible. Fortunately the available debugging tools allow a motivated developer to diagnose and resolve intermittent or sporadic problems.

XP has "structured exception handling", which can produce a minidump or crash dump when Flash Player crashes. This can be analyzed in a post-mortem fashion. Alternatively, a developer can use "just in time" debugging and connect to a customer's computer at the point of failure.

If none of those work, it's easy to produce and distribute an instrumented version of the flash player which gathers additional diagnostic information near the failure point.

I don't work at MS, but these are well-known common techniques. I don't know why this problem has dragged on so long. More info:

http://www.microsoft.com/msj/0197/exception/exception.aspx

http://en.wikipedia.org/wiki/Exception_handling

http://www.dotnetjunkies.com/Tutorial/030F9042-C48A-4210-9977-0425C231BF75.dcik

http://www.codeproject.com/debug/postmortemdebug_standalone1.asp?df=100&forumid=3419&exp=0&select=1000980&fr=26

http://www.compuware.com/products/devpartner/visualc.htm

I finally hit a site that insists on V9 (I don't know if it actually needs it, but it does demand it :).

www.ElfYourSelf.com.

I'm really at a loss to miss this crucial bit of nonsense. (It's actually a viral marketing thing for OfficeMax).

But I will continue to wait for a fix. Not an "almost fix" or a "don't do that in IE" fix, or "try and keep your number of tabs small" fix, but simply a fix fix. (Not the attempts to desparately point fingers at others as in, "well, uh, this is really an IE/XP/User/Hardware bug, though we think that once in a great while our software might be ever so slightly broken though we don't know WHY people are making such a fuss over it. Geesh, doesn't everyone's product do this?".

My sarcasm is not meaningless. It's an attempt to get Adobe to WAKE UP, though if experience is a guide, it will be fruitless too. From a simple user's point of view the fact is simply this: Adobe Flash is broken, has been broken for quite a while, and is likely to remain broken". In a corporate environment, we block our users from installing/upgrading to V9. Who needs the internal help desk calls about IE crashes?".

- Greg

On that note I have a ticket number that I received two weeks ago and still have not heard back from anyone in tech support.

As SpoiledDiegan suggested, Im wondering how long most of are going to fight the urge to simply dump IE and switch over to Mozilla Firefox, which, from what I hear, is superior in so many ways to IE 7, even aside from this flash9.ocx problem.

Spoiled..

"R6025
- pure virtual function call" is actually a different issue. This is a Flash Player issue which was introduce when the current version of the Flash Player 9 Update went live (9.0r64). It's been repaired internally and won't happen when FP9 Update finally goes live.

Okay, I'm back to this forum with some new info. The crash hasn't gone away and I've done everything from this forum to as many as I can find across the web. The latest website was interesting that was crashing on me, Citicards(citibanks) website. I can log in fine, but when I go to view statements or history or what have you, it crashes consistently. So I disabled the plugin and got in fine, checked what I needed and then decided to enable that plugin again before I left, see what that would do. I refreshed the page and tried to browse to the next page and instead of MS's worthless crash error, I finally got something different. I'm not sure if I can post an image in here, so I'll just write it out:

Microsoft Visual C++ Runtime Library
Runtime Error!
Program: C:\Program Files (x86)\Internet Explorer\iexplore.exe

R6025
- pure virtual function call

And you know what, Microsoft has lots of literature on this error, though if you search R6025 and iexplore.exe you get nothing on their website. From what I found though, one of the causes is bad programming.

This sounds like something that I've been feeling it has been since the beginning. I'm not sure if non-Vista users are also getting the same type of crash, but it also seems linked to maybe the release of IE7 if it does go back that far. I can clear my cache/history and get a little farther on a website each time, but when I revisit a page that cached a flash video, ad, you name it, I'm pretty sure that is when it crashes. I think it works fine in some areas where flash programmers may have figured out how Vista/IE7 works, but I think MS's side is to blame. They've "reprogrammed" their stuff so much, I don't blame most for not being able to keep up. I'm going to test some things out, deleting my history on each page before I go to the next, see how that one goes. I am on Vista Ultimate 64-bit, but I haven't had much problems except this flash.ocx thing so far. MS is a piece of crap, we really need someone to "revolutionize" this industry, Leopard is nice, 20X better than Vista, with some actual improvements. Maybe Apple will open some doors and put the monopoly on its face. Course I'm sure Leopard will just cause some new problems, but what else is new.

Any kind of feeback would be appreciated. Back to testing some more things and painfully surfing the internet.