Closed Bug 109719 Opened 24 years ago Closed 24 years ago

Mozilla Fails Browser crash test

Categories

(Core :: Layout, defect)

x86
Windows ME
defect
Not set
critical

Tracking

()

RESOLVED DUPLICATE of bug 77606

People

(Reporter: quattro, Assigned: attinasi)

References

()

Details

(Keywords: crash, hang)

As i was wanedring around internet i foud some interesting page (look URL above). It looks lige guys are searching for known crashers and apply them one by one to your browser. It would be fun to say Moz passes this one, also, i believe it can be a good tool for testing. build 2001110803 crashes at sixth or seventh test. IE 5.5 crashes at fourth:)
2001111108/WinNT4 - Mozilla hang, only page title displayed in Mozilla window caption (and still eating memory). Key: +crash, +hang
Severity: normal → critical
Keywords: crash, hang
2001111203/WinNT4 hangs too.
Build ID: 2001 11 08 03, hangs and sucks all CPU on Windows 2000
-> Layout (?) I get a hang with win2k build 20011111..
Assignee: asa → attinasi
Status: UNCONFIRMED → NEW
Component: Browser-General → Layout
Ever confirmed: true
QA Contact: doronr → petersen
Just adding me to CC ... WIn98SE and Moz (201110903) crash also
Ha - I've had this one before. This is the killer: <SCRIPT> var color = new Array; color[1] = "white"; color[2] = "white"; for(x = 0; x <3; x++) { document.bgColor = color[x] if(x == 2) { x = 0; } } </SCRIPT> So, somebody writes an infinite loop in JS and complains that it hangs the browser - what do they expect?!? Marking INVALID
Status: NEW → RESOLVED
Closed: 24 years ago
Resolution: --- → INVALID
Attinasi : I agree that this is not Mozilla's fault ! But can we not have an emergency exit if we loop in JS ? This page works in NS4.77 and every Page can easy kill Mozilla/NS6.x and you will loose all open browser windows....
yes, the page loads completly and moz passes all tests after diabling javascript. BUT. user does not know that. it seems to him browser has died. so it is more of security question. should we allow such code? this is after all some kind of denial of service (browser stops responding, cpu goes to 100%...) isn't it possible to DETECT an infinite loop in source before actual rendering and pop a warning?
this is a dup
Status: RESOLVED → REOPENED
Resolution: INVALID → ---
*** This bug has been marked as a duplicate of 77606 ***
Status: REOPENED → RESOLVED
Closed: 24 years ago24 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.