User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.0.13) Gecko/2009073022 Firefox/3.0.13 Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.0.13) Gecko/2009073022 Firefox/3.0.13 As you can see in Build Identifier i had to "downgrade" firefox to version 3.0.13 wich is the latest version that can print webpages with my Canon LBP800. Both 3.5.2 and 3.5.3 crashes when i try to print pages from the browser. I paste my crashreport here for you to see. Signature CAPRENN.DLL@0x672f UUID 516acab7-6f82-4c24-9110-e0b162090910 Time 2009-09-10 11:31:08.204203 Uptime 23 Last Crash 607205 seconds before submission Product Firefox Version 3.5.3 Build ID 20090824101458 OS Windows NT OS Version 5.1.2600 Service Pack 3 CPU x86 CPU Info AuthenticAMD family 15 model 31 stepping 0 Crash Reason EXCEPTION_ACCESS_VIOLATION Crash Address 0x8cfb8a0 User Comments Processor Notes Reproducible: Always Steps to Reproduce: 1. open a web page no matter wich 2. push print button or chose print from filemenu 3. webbrowser crashes almost instantly Actual Results: bugreport windows comes up and i am suggested to report the bug and then restart firefox Expected Results: i should have print out the webpage that i wanted

Confirmed based on multiple bug reports. See also bug 517233 for a similar crash in Canons print driver. Reporter, which version of the driver you have installed?

I installed v1.05, which is only available on the Asian-Canon-sites. This was the only Driver that works on XP *and* Vista in 2007. Meanwhile version R1.10 for Vista exists, but I never used it.

I´ve used the latest available from canon´s Swedish website Version 1.00.1.005. Yes i know it´s not official supporting winXP but have worked just fine until this release of firefox 3.5.2/3.5.3

Peter, which revision of that driver you have installed? There are listed R1.11 and R1.12 on the drivers page here: http://software.canon-europe.com/products/0000551.asp Ruediger, when did it started to happen for you? With which version of Firefox (SeaMonkey) you don't see this problem?

(In reply to comment #2) > I installed v1.05, which is only available on the Asian-Canon-sites. > Meanwhile version R1.10 for Vista exists, but I never used it. Forgot to say, that my printer is a Canon LBP-810. I now tested the driver v. R1.10 on my Win-XP-machine and Vista-machine, and the all crashes disappeared in FF3.55, Namoroka and also SM-nightly.

(In reply to comment #4) > Peter, which revision of that driver you have installed? There are listed R1.11 > and R1.12 on the drivers page here: This are drivers for 'older than XP'-Windows.

What about those R1.12 drivers? Are those older or do newer drivers crash?

(In reply to comment #7) > What about those R1.12 drivers? Are those older or do newer drivers crash? It looks like this drivers from 2003 are the latest for the reporters printer. They are actual build for Win 2k and but also run on XP. My R1.10 is from 2007, but it is not for the Reporters LBP-800 printer.

Peter, can you please check the link in my comment 4 if those are more recent versions of the driver?

If you do select OS in that window you schould see that the latest driver is the one i´ve posted earlier. 1. Canon CAPT Printer Driver R1.11 (1.00.1.005) Canon Advanced Printing Technology Driver R1.11 No matter which OS you choose the driver version is the same. And fortunately it is´nt work with ff. Btw I´ve choose to change my printer to another Canon LBP3370 and it works perfectly with Win XP/7 64 bit and Linux Over TCP/IP NOP at all. I was very found of my old Canon but somewhere on the road to perfection we grow apart. Canon LBP 800 RIP. For me this case is closed.

Sorry i meant fortunately. Not better in English than my spell checker.

You see it got wrong again damn MS Unfortunately that is.

Ok, sad to hear that you cannot give more information from now on. The next step would be to check builds of Firefox which are produced on a nightly basis. In that way we could identify the day when it happened for the first time with an early Firefox 3.5. The downside is that it will take time to test this. On the other hand we could say afterward if this is a regression on our side or if we have fixed a bug on our side which let the driver to malfunction. If one of you is interested in this work please comment on this bug. Thanks.

(In reply to comment #13) > The next step would be to check builds of Firefox which are produced on a > nightly basis. > If one of you is interested in this work please comment on this bug. Thanks. Not interested, but someone has to do the job :-) Okay Henrik, you know, I'm a SeaMonkey-guy cause thats the better product :-) last good in my archiv: Trunk-SM 20090717070233 first bad in my archiv: Trunk-SM 20090718092218 Happy bug-hunting :-)

Ruediger, many thanks! Can you please open about:buildconfig and copy&paste the link for the changeset here? That's needed to get closer.

For Trunk-SM 20090718092218: http://hg.mozilla.org/mozilla-central/rev/38bc9063e689

Sorry for the stress. :) But what's the changeset for the 20090717070233 build?

I cannot see any suspicious checkin on that day: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=b4e7f7d88a8c&tochange=38bc9063e689 Ruediger, does it happen whether which web page you have open? Even with only one tab and about:blank? Ruediger, can you add one of your crash reports please?

Crash on every site. Even on a single about:blank http://crash-stats.mozilla.com/report/index/1dae271a-386c-4f50-935f-83d562091124

Bug 479667 (enabling DEP) is the only thing that strikes me as likely in that pushlog -- that change is specifically about turning suspicious activity into crashes. However, ted doesn't think we'd be getting crash reports from DEP crashes, which would suggest that it's not that... but nothing else really looks suspicious. Ruediger, I'm going to generate a few builds taken from intermediate points during the range from comment 19. I'll post links here when they're done -- if you could give them a shot & report back, it'd be very helpful for tracking down what caused this. (Thanks for all your testing thus far, too!)

Actually, Ruediger: Can you test these two builds: http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/2009/07/2009-07-21-04-mozilla-1.9.1/ http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/2009/07/2009-07-22-04-mozilla-1.9.1/ Those are the nightly builds of Gecko 1.9.1 (Firefox 3.5), before & after Bug 479667 landed there. I predict that the first will be fine, and the second will crash -- let me know if that's correct. Pushlog between those builds is pretty small: http://hg.mozilla.org/releases/mozilla-1.9.1/pushloghtml?fromchange=d1c4c407468c&tochange=288184ebf470 ...so if this breaks between those nightlies, then this is almost certainly Bug 479667

Bingo! First print, second crash. bb6c501d-c805-49a6-acce-577a02091124

Great, thanks! I'm confident saying that this started crashing due to bug 479667, then, because: - That's the only checkin in common between the 1.9.1 and m-c pushlogs - That's the only suspicious-looking checkin (for a crash in a Canon printer driver) in either pushlog So the upshot of this is: this crash is due to a bug (or bad design decision) in the Canon printer driver, which is exposed by Firefox, because Firefox enables DEP to protects against exploits from this class of bugs.[1] Is there a newer version of the driver available? I think the best solution would be to update to a fixed driver version, assuming one is available... [1] http://en.wikipedia.org/wiki/Data_Execution_Prevention

(In reply to comment #21) > Bug 479667 (enabling DEP) is the only thing that strikes me as likely in that > pushlog -- that change is specifically about turning suspicious activity into > crashes. My wife told me, that trying to print from notepad.exe also caused crashes, and brings up the DEP-Warning-Popup from Windows before the crash. I tested that now, and she is right. I saved the windows-crashreport, if you need it.

(In reply to comment #24) > Is there a newer version of the driver available? I think the best solution > would be to update to a fixed driver version, assuming one is available... Not for the reporters printer, but for me, and that works with the newer mozillas.

Yeah -- according to comment 5 (and confirmed via a glance at the link in comment 4), there's no later version of the Canon Driver for the reporter's LBP-800 printer. He's using the latest version, which is from 2003 and which doesn't mention supporting Windows XP. Resolving as INVALID, since this is not a Firefox bug. For Peter (reporter) and others who are affected by this and can't update your driver: You should be able to work around it by disabling DEP in Windows XP Control Panel (either globally, or just for Firefox). Rough steps to do this are here: http://en.wikipedia.org/wiki/Data_Execution_Prevention#Configuration http://support.microsoft.com/kb/875352 (or you might be able to find better directions with some Googling) I can't provide detailed steps, as I don't use Windows and haven't had to disable DEP before, but those pages should hopefully have all the information. If anyone (Ruediger? Peter?) can confirm that disabling DEP works around this issue (and ideally provide instructions for others to follow), that would be awesome...

Peter and especially Ruediger, many thanks for helping us to investigate this crash.

Ruediger Lahl: we can't really use it, but you really should contact Canon and complain. They *should* be able to use that .dmp. If you get a case number of some sorts, we wouldn't mind having it here so that other victims of their crashing driver who visit this bug can get your case number and use it when they complain to Canon. called tech support. they suggest that i use a pen and piece of paper: Canon RCC Unit 130 Centenial Avenue Elstreethill South Borehamwood Herdshire WD63SE England