Closed Bug 154699 (OSX_Java) Opened 23 years ago Closed 22 years ago

Mozilla crashes when launching Java applets in Jaguar [@ JVM_RegisterSignal]

Categories

(Core Graveyard :: Java: OJI, defect, P1)

Product:
Core Graveyard
Component:
Java: OJI
Platform:
PowerPC
macOS
Type:
defect

Tracking

(Not tracked)

Status:
VERIFIED FIXED
Milestone:
mozilla1.0.1

People

(Reporter: vduvvuri, Assigned: joe.chou)

References

Details

(Keywords: crash, Whiteboard: [adt1 RTM] [ETA 08/20])

Crash Data

Attachments

(1 file, 1 obsolete file)

patch to clear exceptions from missing JSObject constructor/field
1.56 KB, patch
sdagley
: review+
bryner
: superreview+
chofmann
: approval+
Details | Diff | Splinter Review
From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.14; Mac_PowerPC) BuildID: 2002061014 I downloaded Netscape pluggin for Java and installed it. Launch applets . Mozilla and Netscape Crashes. Make sure that /library/Internet PLugins/ MRJPluginCarbon is the only plugin available. The crash log suggests that the VM is not able to spawn. Reproducible: Always Steps to Reproduce: 1.Download MRJPlugin from http://homepage.mac.com/pcbeard/ MRJPlugin/ 2. Launch Applet like http://segal.org/PerkyChoice/index.html 3. Notice that Netscape 7.0 and Mozilla 1.1 Alpha crashes. Actual Results: Crash Expected Results: Should work fine Date/Time: 2002-06-27 15:02:56 -0700 OS Version: 10.2 (Build 6C35) Host: duvvve.apple.com Command: Mozilla PID: 477 Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000030 Thread 0: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991f8d14 in JVM_RegisterSignal #3 0x991fe13c in JVM_RegisterSignal #4 0x991e0a48 in JVM_RegisterSignal #5 0x99210b10 in JVM_RegisterSignal #6 0x9922cd14 in JVM_RegisterSignal #7 0x9924fcf4 in JVM_RegisterSignal #8 0x991e3590 in JVM_RegisterSignal #9 0x9915155c in JVM_RegisterSignal #10 0x9914d2c8 in JVM_RegisterSignal #11 0x990088bc in JVM_GetMethodIxByteCodeLength #12 0x7a221848 in LockAcquire #13 0x7a221030 in initialize(JNIEnv_*, bool) #14 0x7ed43460 in InitJavaEmbedding(JNIEnv_*&) #15 0x7ed457f4 in CreateJavaApplet #16 0x0321f734 in MRJContext::loadApplet(void) #17 0x03220298 in MRJContext::setWindow(nsPluginWindow *) #18 0x0321bea4 in MRJPluginInstance::SetWindow(nsPluginWindow *) #19 0x031b5fec in InstantiateEmbededPlugin__16nsPluginHostImplFPCcP6nsIURIP22n sI #20 0x0351f794 in InstantiatePlugin__13nsObjectFrameFP14nsIPresContextR19nsHTM LR #21 0x0351e4a0 in 0x351e4a0 #22 0x035537c0 in 0x35537c0 #23 0x0354c828 in 0x354c828 #24 0x0354c518 in DoReflowInlineFrames__12nsBlockFrameFR18nsBlockReflowStateR1 2n #25 0x0354c188 in nsBlockFrame::DoReflowInlineFramesAuto(nsBlockReflowState &) #26 0x0354bf08 in ReflowInlineFrames__12nsBlockFrameFR18nsBlockReflowState19ns Li #27 0x0354a318 in ReflowLine__12nsBlockFrameFR18nsBlockReflowState19nsLineList _i #28 0x03549904 in 0x3549904 #29 0x03547d2c in nsBlockFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics & ) #30 0x0356fdb8 in DoReflowBlock__20nsBlockReflowContextFR17nsHTMLReflowState14 ns #31 0x0356f818 in ReflowBlock__20nsBlockReflowContextFP8nsIFrameRC6nsRectiR18n sC #32 0x0354b888 in 0x354b888 #33 0x03549f3c in 0x3549f3c #34 0x03549904 in 0x3549904 #35 0x03547d2c in nsBlockFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics & ) #36 0x0356fdb8 in DoReflowBlock__20nsBlockReflowContextFR17nsHTMLReflowState14 ns #37 0x0356f818 in ReflowBlock__20nsBlockReflowContextFP8nsIFrameRC6nsRectiR18n sC #38 0x0354b888 in 0x354b888 #39 0x03549f3c in 0x3549f3c #40 0x03549904 in 0x3549904 #41 0x03547d2c in nsBlockFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics & ) #42 0x034a546c in ReflowChild__16nsContainerFrameFP8nsIFrameP14nsIPresContextR 19 #43 0x036449bc in CanvasFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &, &) #44 0x03601a00 in Reflow__19nsBoxToBlockAdaptorFR16nsBoxLayoutStateP14nsIPresC on #45 0x036013e4 in nsBoxToBlockAdaptor::DoLayout(nsBoxLayoutState &) #46 0x035fbfec in nsBox::Layout(nsBoxLayoutState &) #47 0x0362dd9c in 0x362dd9c #48 0x035fbfec in nsBox::Layout(nsBoxLayoutState &) #49 0x03604a08 in LayoutChildAt__14nsContainerBoxFR16nsBoxLayoutStateP6nsIBoxR C6 #50 0x035f04dc in nsGfxScrollFrameInner::LayoutBox(nsBoxLayoutState &, nsIBox *) #51 0x035f09bc in 0x35f09bc #52 0x035f0594 in nsGfxScrollFrame::DoLayout(nsBoxLayoutState &) #53 0x035fbfec in nsBox::Layout(nsBoxLayoutState &) #54 0x035bd1e4 in nsBoxFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &, const &) #55 0x035ef5d0 in Reflow__16nsGfxScrollFrameFP14nsIPresContextR19nsHTMLReflowM et #56 0x034a546c in ReflowChild__16nsContainerFrameFP8nsIFrameP14nsIPresContextR 19 #3 0x9054ca70 in CarbonConditionWait #4 0x9054e720 in CarbonInetOperThreadFunc #4 0x9922c43c in JVM_RegisterSignal #5 0x9922c958 in JVM_RegisterSignal #6 0x9922c618 in JVM_RegisterSignal #4 0x9903d69c in JVM_RegisterSignal #5 0x9911d448 in JVM_RegisterSignal #6 0x99011860 in JNI_CreateJavaVM_Impl #7 0x90021f6c in _pthread_body Thread 12: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991f8d14 in JVM_RegisterSignal #3 0x99210b3c in JVM_RegisterSignal #4 0x9915274c in JVM_RegisterSignal #5 0x9916af70 in JVM_RegisterSignal #6 0x9911d448 in JVM_RegisterSignal #7 0x99011860 in JNI_CreateJavaVM_Impl #8 0x90021f6c in _pthread_body PPC Thread State: srr0: 0x99217548 srr1: 0x0000d030 vrsave: 0x00000000 xer: 0x00000000 lr: 0x991ff0c0 ctr: 0x9925f750 mq: 0x00000000 r0: 0x00000002 r1: 0xf0386a50 r2: 0x991ff0c0 r3: 0x6c0c8178 r4: 0x6c0c8178 r5: 0x00000000 r6: 0x6e000000 r7: 0xa903a2d8 r8: 0xa8ffeebc r9: 0x00000000 r10: 0x00000000 r11: 0x00000008 r12: 0x9925f750 r13: 0xf0386b80 r14: 0x00000000 r15: 0xf0386ba0 r16: 0xa8fff37c r17: 0xa8fff37c r18: 0xa8fff37c r19: 0x00000003 r20: 0x00000001 r21: 0x6c0c8200 r22: 0x0003ffb8 r23: 0x00000000 r24: 0x04153990 r25: 0xa9001e00 r26: 0x04153990 r27: 0x6c0c8178 r28: 0x6c500000 r29: 0x6c0c8178 r30: 0xa8ffeebc r31: 0x991feebc **********
Venkata, I tried on Mac os 10.1.5 (netscape branch build: 2002-06-27-05-1.0.0) MRJ: v1.0, it loads fine without crashing.
I am ccing Patrick Beard and Brian Neese
No I am trying on a more recent version of Max OS 10.2 (Jaguar 6c77). Let me try on on 10.1.5 also. Will let u know my results.
Tried on 10.1.5 . Still Crashes. Make sure that u have only MRJPluginCarbon.plugin in the /library/Internet Plugins/ directory
Re-assign to Patrick Beard.
Assignee: joe.chou → beard
WorksForMe using FizzillaCFM/2002062203. Venkata, could you attach the entire crash report? The one you included only shows threads 0 and 12. I've noticed with Java crashes it's sometimes another thread that actually crashed.
Keywords: crash
Summary: Netscape 7.0 and Mozilla crashes when launching applets using MRJPlugin provided by Netscape → Mozilla crashes when launching Java applets
Thats all the threads that are shown in the Crash Log.
Looks like there was some cut copy error. Here is the crash log again... Date/Time: 2002-07-01 11:00:33 -0700 OS Version: 10.2 (Build 6C35) Host: duvvve.apple.com Command: Mozilla PID: 3786 Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000030 Thread 0: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991f8d14 in JVM_RegisterSignal #3 0x991fe13c in JVM_RegisterSignal #4 0x991e0a48 in JVM_RegisterSignal #5 0x99210b10 in JVM_RegisterSignal #6 0x9922cd14 in JVM_RegisterSignal #7 0x9924fcf4 in JVM_RegisterSignal #8 0x991e3590 in JVM_RegisterSignal #9 0x9915155c in JVM_RegisterSignal #10 0x9914d2c8 in JVM_RegisterSignal #11 0x990088bc in JVM_GetMethodIxByteCodeLength #12 0x7a221848 in LockAcquire #13 0x7a221030 in initialize(JNIEnv_*, bool) #14 0x7ed43460 in InitJavaEmbedding(JNIEnv_*&) #15 0x7ed457f4 in CreateJavaApplet #16 0x031f9734 in MRJContext::loadApplet(void) #17 0x031fa298 in MRJContext::setWindow(nsPluginWindow *) #18 0x031f5ea4 in MRJPluginInstance::SetWindow(nsPluginWindow *) #19 0x031b7fec in InstantiateEmbededPlugin__16nsPluginHostImplFPCcP6nsIURIP22nsI #20 0x034ca794 in InstantiatePlugin__13nsObjectFrameFP14nsIPresContextR19nsHTMLR #21 0x034c94a0 in 0x34c94a0 #22 0x034fe7c0 in 0x34fe7c0 #23 0x034f7828 in 0x34f7828 #24 0x034f7518 in DoReflowInlineFrames__12nsBlockFrameFR18nsBlockReflowStateR12 n #25 0x034f7188 in nsBlockFrame::DoReflowInlineFramesAuto(nsBlockReflowState &) #26 0x034f6f08 in ReflowInlineFrames__12nsBlockFrameFR18nsBlockReflowState19nsLi #27 0x034f5318 in ReflowLine__12nsBlockFrameFR18nsBlockReflowState19nsLineList_i #28 0x034f4904 in 0x34f4904 #29 0x034f2d2c in nsBlockFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &) #30 0x0351adb8 in DoReflowBlock__20nsBlockReflowContextFR17nsHTMLReflowState14n s #31 0x0351a818 in ReflowBlock__20nsBlockReflowContextFP8nsIFrameRC6nsRectiR18ns C #32 0x034f6888 in 0x34f6888 #33 0x034f4f3c in 0x34f4f3c #34 0x034f4904 in 0x34f4904 #35 0x034f2d2c in nsBlockFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &) #36 0x0351adb8 in DoReflowBlock__20nsBlockReflowContextFR17nsHTMLReflowState14n s #37 0x0351a818 in ReflowBlock__20nsBlockReflowContextFP8nsIFrameRC6nsRectiR18ns C #38 0x034f6888 in 0x34f6888 #39 0x034f4f3c in 0x34f4f3c #40 0x034f4904 in 0x34f4904 #41 0x034f2d2c in nsBlockFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &) #42 0x0345046c in ReflowChild__16nsContainerFrameFP8nsIFrameP14nsIPresContextR1 9 #43 0x035ef9bc in CanvasFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &, &) #44 0x035aca00 in Reflow__19nsBoxToBlockAdaptorFR16nsBoxLayoutStateP14nsIPresCon #45 0x035ac3e4 in nsBoxToBlockAdaptor::DoLayout(nsBoxLayoutState &) #46 0x035a6fec in nsBox::Layout(nsBoxLayoutState &) #47 0x035d8d9c in 0x35d8d9c #48 0x035a6fec in nsBox::Layout(nsBoxLayoutState &) #49 0x035afa08 in LayoutChildAt__14nsContainerBoxFR16nsBoxLayoutStateP6nsIBoxRC6 #50 0x0359b4dc in nsGfxScrollFrameInner::LayoutBox(nsBoxLayoutState &, nsIBox *) #51 0x0359b9bc in 0x359b9bc #52 0x0359b594 in nsGfxScrollFrame::DoLayout(nsBoxLayoutState &) #53 0x035a6fec in nsBox::Layout(nsBoxLayoutState &) #54 0x035681e4 in nsBoxFrame::Reflow(nsIPresContext *, nsHTMLReflowMetrics &, const &) #55 0x0359a5d0 in Reflow__16nsGfxScrollFrameFP14nsIPresContextR19nsHTMLReflowMe t #56 0x0345046c in ReflowChild__16nsContainerFrameFP8nsIFrameP14nsIPresContextR1 9 #57 0x035382f0 in Reflow__13ViewportFrameFP14nsIPresContextR19nsHTMLReflowMetric #58 0x03464e98 in Dispatch__17IncrementalReflowFP14nsIPresContextR19nsHTMLReflow #59 0x03475ec4 in PresShell::ProcessReflowCommands(int) #60 0x03475908 in HandlePLEvent(ReflowEvent *) #61 0x0135b270 in PL_HandleEvent #62 0x0135b0dc in PL_ProcessPendingEvents #63 0x013010fc in nsEventQueueImpl::ProcessPendingEvents(void) #64 0x0291bb9c in nsMacNSPREventQueueHandler::ProcessPLEventQueue(void) #65 0x0291ba40 in nsMacNSPREventQueueHandler::RepeatAction(EventRecord const &) #66 0x0255eb14 in Repeater::DoRepeaters(EventRecord const &) #67 0x02931f58 in nsMacMessagePump::DispatchEvent(int, EventRecord *) #68 0x02931c80 in nsMacMessagePump::DoMessagePump(void) #69 0x029315fc in nsAppShell::Run(void) #70 0x02856afc in nsAppShellService::Run(void) #71 0x0122ef1c in main1(int, char **, nsISupports *) #72 0x0122f95c in main Thread 1: #0 0x90004bec in syscall #1 0x9054683c in BSD_waitevent #2 0x905456bc in CarbonSelectThreadFunc #3 0x90021f6c in _pthread_body Thread 2: #0 0x90041168 in semaphore_wait_signal_trap #1 0x90040f38 in _pthread_cond_wait #2 0x9054ca90 in BSD_pthread_cond_wait #3 0x9054ca70 in CarbonConditionWait #4 0x9054a7c0 in CarbonOperationThreadFunc #5 0x90021f6c in _pthread_body Thread 3: #0 0x90045048 in semaphore_timedwait_signal_trap #1 0x90040e50 in _pthread_cond_wait #2 0x9026423c in TSWaitOnSemaphoreCommon #3 0x90298750 in TimerThread #4 0x90021f6c in _pthread_body Thread 4: #0 0x90041168 in semaphore_wait_signal_trap #1 0x90040f38 in _pthread_cond_wait #2 0x9026431c in TSWaitOnSemaphoreCommon #3 0x902732a8 in AsyncFileThread(void*) #4 0x90021f6c in _pthread_body Thread 5: #0 0x90041168 in semaphore_wait_signal_trap #1 0x90040f38 in _pthread_cond_wait #2 0x9054ca90 in BSD_pthread_cond_wait #3 0x9054ca70 in CarbonConditionWait #4 0x9054e720 in CarbonInetOperThreadFunc #5 0x90021f6c in _pthread_body Thread 6: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991f8e5c in JVM_RegisterSignal #3 0x991e5a98 in JVM_RegisterSignal #4 0x99011860 in JNI_CreateJavaVM_Impl #5 0x90021f6c in _pthread_body Thread 7 Crashed: #0 0x99217548 in JVM_RegisterSignal #1 0x991ff0c0 in JVM_RegisterSignal #2 0x9924f700 in JVM_RegisterSignal #3 0x991660a8 in JVM_RegisterSignal #4 0x9922c43c in JVM_RegisterSignal #5 0x9922c958 in JVM_RegisterSignal #6 0x9922c618 in JVM_RegisterSignal #7 0x99011860 in JNI_CreateJavaVM_Impl #8 0x90021f6c in _pthread_body Thread 8: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991750a0 in JVM_RegisterSignal #3 0x991d5e30 in JVM_RegisterSignal #4 0x99004c00 in JVM_MonitorWait #5 0x04f0b668 in 0x4f0b668 #6 0x04f0968c in 0x4f0968c #7 0x04f0968c in 0x4f0968c #8 0xa9030428 in vtable for __cxxabiv1::__class_type_info #9 0x9924dbbc in JVM_RegisterSignal #10 0x9924e0b4 in JVM_RegisterSignal #11 0x9924e120 in JVM_RegisterSignal #12 0x9903a90c in JVM_RegisterSignal #13 0x9911d448 in JVM_RegisterSignal #14 0x99011860 in JNI_CreateJavaVM_Impl #15 0x90021f6c in _pthread_body Thread 9: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991750a0 in JVM_RegisterSignal #3 0x991d5e30 in JVM_RegisterSignal #4 0x99004c00 in JVM_MonitorWait #5 0x04f0b668 in 0x4f0b668 #6 0x04f0968c in 0x4f0968c #7 0x04f095cc in 0x4f095cc #8 0x04f095cc in 0x4f095cc #9 0xa9030428 in vtable for __cxxabiv1::__class_type_info #10 0x9924dbbc in JVM_RegisterSignal #11 0x9924e0b4 in JVM_RegisterSignal #12 0x9924e120 in JVM_RegisterSignal #13 0x9903a90c in JVM_RegisterSignal #14 0x9911d448 in JVM_RegisterSignal #15 0x99011860 in JNI_CreateJavaVM_Impl #16 0x90021f6c in _pthread_body Thread 10: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991f8d78 in JVM_RegisterSignal #3 0x991f87a0 in JVM_RegisterSignal #4 0x991915a8 in JVM_RegisterSignal #5 0x99011860 in JNI_CreateJavaVM_Impl #6 0x90021f6c in _pthread_body Thread 11: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x99210b3c in JVM_RegisterSignal #3 0x9903db20 in JVM_RegisterSignal #4 0x9903d69c in JVM_RegisterSignal #5 0x9911d448 in JVM_RegisterSignal #6 0x99011860 in JNI_CreateJavaVM_Impl #7 0x90021f6c in _pthread_body Thread 12: #0 0x90078a88 in mach_msg_trap #1 0x90005a70 in mach_msg #2 0x991f8d14 in JVM_RegisterSignal #3 0x99210b3c in JVM_RegisterSignal #4 0x9915274c in JVM_RegisterSignal #5 0x9916af70 in JVM_RegisterSignal #6 0x9911d448 in JVM_RegisterSignal #7 0x99011860 in JNI_CreateJavaVM_Impl #8 0x90021f6c in _pthread_body PPC Thread State: srr0: 0x99217548 srr1: 0x0000d030 vrsave: 0x00000000 xer: 0x00000000 lr: 0x991ff0c0 ctr: 0x9925f750 mq: 0x00000000 r0: 0x00000002 r1: 0xf0386a50 r2: 0x991ff0c0 r3: 0x6c0c8178 r4: 0x6c0c8178 r5: 0x00000000 r6: 0x6e000000 r7: 0xa903a2d8 r8: 0xa8ffeebc r9: 0x00000000 r10: 0x00000000 r11: 0x00000008 r12: 0x9925f750 r13: 0xf0386b80 r14: 0x00000000 r15: 0xf0386ba0 r16: 0xa8fff37c r17: 0xa8fff37c r18: 0xa8fff37c r19: 0x00000003 r20: 0x00000001 r21: 0x6c0c8200 r22: 0x0003ffb8 r23: 0x00000000 r24: 0x04132990 r25: 0xa9001e00 r26: 0x04132990 r27: 0x6c0c8178 r28: 0x6c500000 r29: 0x6c0c8178 r30: 0xa8ffeebc r31: 0x991feebc
Summary: Mozilla crashes when launching Java applets → Mozilla crashes when launching Java applets [@ JVM_RegisterSignal]
Using a 6/26 build of Netscape 7 and running with the MRJ Plug-in in either /Library/Internet Plug-ins or installed as part of the Netscape package this WFM on 10.1.5.
Also WorksForMe using FizzillaCFM/2002062203. Venkata, are you saying that "MRJPluginCarbon" should be the only plug-in loaded by Mozilla? That is, only "OJI Plugin for Mac OS X, v1.0" should show up under [about:plugins]?
I downloaded the latest build of Mozilla Build 2002070808 I even tried again with Netscape 7.0 PR1 Build 5/22 My Mac OS X 10.1.5 Build 5560 1. The crash happens everytime when i try to launch the applets. Note that here the MRJPluginCarbon.plugin is located in /Library/Internet Plugins/ directory 2. When i clicked on "About Plugins" Menu item, it does show up as Plugin for Mozilla and Netscape as "OJI Plugin for Mac OS X, v1.0"
Is it possible that you have the MRJ plugin installed in both the package (where it resides by default) AND in /Library/Internet Plug-ins?
Yes MRJPlugin is installed in both the default package and /Library/ Internet Plugins/
I downloaded the plugin from http://homepage.mac.com/pcbeard/MRJPlugin/ and installed it in /Library/Internet Plug-ins and ran NS 7 with both plug-ins (the pre-packaged one and the downloaded one) installed. Entering about:plugins in the url bar shows that the one in the Netscape Package is loaded, and I experienced no crashes. This is still WFM.
Worked fine for me on another machine (10.15 Puma). However, Problems with 10.2 Jaguar. Doing more testing on this.
Using 1.1b or Chimera 0.4, and latest Jaguar (106), I get the following crashed thread any time I try to access any applet (i.e. www.access-board.gov): The crashed thread seems to always look like this for me: Thread 7 Crashed: #0 0x938ea698 in eh_rest_world_r10 #1 0x93787ab8 in JVM_GetMethodIxLocalsCount #2 0x9378c1a8 in JVM_GetCPMethodNameUTF #3 0x937c1874 in JVM_MonitorNotifyAll #4 0x937b6eec in JVM_SetThreadPriority #5 0x937949e4 in JVM_Send #6 0x938311cc in JNI_CreateJavaVM_Impl #7 0x937d527c in JVM_GetClassMethodsCount #8 0x90021428 in _pthread_body Probably unrelated, Apple's Java Plugin seems to crash Mozilla going to about:plugins. This crash is 100% reproducable for me, with any Applet at all..
Steve, that's a different stack. Please file a separate bug and attach your crash report to it.
I am expiecning the same problems reported by Steve . can someone send me that bug number. It happens with the latest jaguar
As this bug was essentially invalidated under OS 10.1.x, I am simply going to co-opt this into a Plugin crashes under Jaguar bug... I am currently getting mixed signals on this. I have been told (and I'm going to be purposely vauge here because even though Jaguar is no longer a secret, some of the internals haven't actually been announced) that there are certain situations which will cause this crash... So far, my testing has indicated a 100% failure rate, so I need to investigate this further.
Status: UNCONFIRMED → NEW
Ever confirmed: true
Summary: Mozilla crashes when launching Java applets [@ JVM_RegisterSignal] → Mozilla crashes when launching Java applets in Jaguar [@ JVM_RegisterSignal]
I filed this with Apple, so they're aware of it. hypothetically speaking, there may have been a new hypothetical build released yesterday, which I'll hypothetically take a look at.
Hypothetically speaking, there would be no reason to believe this would have gotten any better with the GM candidate. If anyone knows why, hypothetically, CrashReporter logs would not be working, please e-mail me.
Disregard that, same symptom, different stack.
*** Bug 160995 has been marked as a duplicate of this bug. ***
*** Bug 162206 has been marked as a duplicate of this bug. ***
*** Bug 161882 has been marked as a duplicate of this bug. ***
*** Bug 161753 has been marked as a duplicate of this bug. ***
*** Bug 162538 has been marked as a duplicate of this bug. ***
*** Bug 162536 has been marked as a duplicate of this bug. ***
I have learned that we may not be cleaning up after some of our JNI calls which can throw Java exceptions, and the Jaguar JVM seems to be corrupting itself if we make certain JNI calls when exceptions are pending. I'm going to work on a patch to the plugin that checks for exceptions and clears them in a more systematic way.
*** Bug 163036 has been marked as a duplicate of this bug. ***
The cause of the crash is that we are somehow failing to look up the method ID of a private JSObject constructor and field. We fail to clear the resulting Java exception, and then the JVM crashes later on when we try to create an applet. A quick fix is to add code to unconditionally clear exceptions at these call sites. Patch coming up.
Status: NEW → ASSIGNED
Blocks: 143047
Keywords: mozilla1.0.1, nsbeta1+
Whiteboard: [adt1 RTM] [ETA 08/20]
This adds exception cleanup code when the private constructor and field are missing from the JSObject class.
Comment on attachment 95940 [details] [diff] [review] patch to clear exceptions from missing JSObject constructor/field r=sdagley
Attachment #95940 - Flags: review+
Comment on attachment 95940 [details] [diff] [review] patch to clear exceptions from missing JSObject constructor/field sr=bryner
Attachment #95940 - Flags: superreview+
adt1.0.1+ (on ADT's behalf) approval for checkin to the 1.0 branch, pending Drivers' approval. pls check this in asap, then replace the "Mozilla1.0.1" keyword with "fixed1.0.1". thanks!
Keywords: adt1.0.1+
Priority: -- → P1
Target Milestone: --- → mozilla1.0.1
Fix checked in on trunk and 1.0 branch.
Status: ASSIGNED → RESOLVED
Closed: 23 years ago
Resolution: --- → FIXED
Comment on attachment 95940 [details] [diff] [review] patch to clear exceptions from missing JSObject constructor/field a=chofmann for 1.0.1
Attachment #95940 - Flags: approval+
Marking as fixed1.0.1 as this has been checked into the 1.0 branch per Comment #36 From Patrick C. Beard.
Keywords: mozilla1.0.1 → fixed1.0.1
*** Bug 163544 has been marked as a duplicate of this bug. ***
javascript java communication is broken now, it seems. I tried my favorite bank account, and although I could log in, starting the java applet to make some attributions to the welfare of the world failed, with a remark: failed to initialize applet.
Johan: do you have any good reason to believe your problem was caused by this checkin? Are you running OSX, and did the problem start with this change? Does it go away if you back out the change or revert to the previous build? Is there any URL we can test with? (admittedly, this may be a problem due to it being behind a bank login). We need to get a handle on this ASAP. Reopening to cover the regression; if it turns out to be spurious or incorrect please re-close.
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Considering that ftp.mozilla.org has no builds of the branch since Patrick checked in the changes, I think it's pretty unlikely that Johan discovered a regression. I have some pretty good test cases to throw at this, but most likely, if the plugin loads, there's no regression of _this_ bug.
I checked out the latest mozilla trunk ( 2002082003), as I have waited for this bugfix for some time now. I am running Mac OS X build C115 The about-plugin works, the log-in in the bank site works, which invokes a java-applet that crashed mozilla in the previous builds, so I could login just like I already did with Explorer, but the javascript-java communication did not work, according to an error message that the bank writes itself (it's a javascript dialog box). Also, news.bbc.co.uk news applet still does not work (does work in explorer and omniweb)
Ok, the "javascript-java communication" (aka LiveConnect) issue makes sense. The problem, and beard will correct me if I mangle this, is due to Apple's ui.jar file containing old Netscape JS and security classes which are not compatible with class files bundled in the MRJCarbonPlugin. Until such time as Apple removes the classes from their ui.jar file, or someone does it for themselves, LiveConnect will not work. Changing this bug back to FIXED as the change checked in by beard does not cause this problem, it only lets you get far enough along in using Java under Jaguar to see that it exists.
Status: REOPENED → RESOLVED
Closed: 23 years ago → 23 years ago
Resolution: --- → FIXED
So then, would the presence of these old security classes explain this: Behind proxy server and go to: http://java.sun.com/docs/books/tutorial/essential/threads/DiningPhilosophers.html I get in the browser window: java.security.AccessControlExcetpion: access denied(java.net.SocketPermission java.sun.com connect,accept,resolve) and I get in the java console: java.security.AccessControlException: access denied (java.net.SocketPermission java.sun.com connect,accept,resolve) at java.security.AccessControlContext.checkPermission(AccessControlContext.java:267) at java.security.AccessController.checkPermission(AccessController.java:394) at java.lang.SecurityManager.checkPermission(SecurityManager.java:540) at java.net.URLClassLoader$5.run(URLClassLoader.java:458) at java.security.AccessController.doPrivileged(Native Method) at java.net.URLClassLoader.getPermissions(URLClassLoader.java:456) at sun.applet.AppletClassLoader.getPermissions(AppletClassLoader.java:167) at sun.plugin.security.PluginClassLoader.getPermissions(PluginClassLoader.java:107) at java.security.SecureClassLoader.getProtectionDomain(SecureClassLoader.java:157) at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:106) at java.net.URLClassLoader.defineClass(URLClassLoader.java:243) at java.net.URLClassLoader.access$100(URLClassLoader.java:51) at java.net.URLClassLoader$1.run(URLClassLoader.java:190) at java.security.AccessController.doPrivileged(Native Method) at java.net.URLClassLoader.findClass(URLClassLoader.java:183) at sun.applet.AppletClassLoader.findClass(AppletClassLoader.java:124) at sun.plugin.security.PluginClassLoader.findClass(PluginClassLoader.java:269) at java.lang.ClassLoader.loadClass(ClassLoader.java:294) at sun.applet.AppletClassLoader.loadClass(AppletClassLoader.java:107) at java.lang.ClassLoader.loadClass(ClassLoader.java:250) at sun.applet.AppletClassLoader.loadCode(AppletClassLoader.java:463) at sun.applet.AppletPanel.createApplet(AppletPanel.java:581) at sun.plugin.AppletViewer.createApplet(AppletViewer.java:1334) at sun.applet.AppletPanel.runLoader(AppletPanel.java:510) at sun.applet.AppletPanel.run(AppletPanel.java:288) at java.lang.Thread.run(Thread.java:491)
I guess I should have added that comment #45 above was using 2002082003, the latest trunk build on 6C115. Using the 1.1b release, it worked fine on the same OS build (once I removed the MRJCarbon from the Mozilla app bundle). So, I am not inclined to think that is a result of anything with the JDK. Or, am I wrong?
(1) Yes, this is no longer crashed anymore on Mac os 10.2 (jaguar), MRJ: v1.0. (netscape branch build: 2002-08-20-05-1.0). This bug is fixed!!! (2)However, just like you visit this url: http://segal.org/PerkyChoice/index.html it is no longer crashed, but when you click the "BACK" or "FORWARD" buttons, it generates the errors like this in the java console.log "java.lang.NullPointerException".... or (3) type: about:plugins or go to HELP > About Plug-ins the errors also occur in the java console. log as well. The logs will attach. This should be filed as a separate bug.
Status: RESOLVED → VERIFIED
Keywords: verified1.0.1
Please refers this bug for details: http://bugzilla.mozilla.org/show_bug.cgi?id=163684
I'm not quite sure that's the right place to say it because it seems to plague Chimera only, but I've seen lots of reports saying that Chimera keeps on crashing when launching a java-enabled page, even with 0819+ builds. It works OK for me, but on other Jag machines (I've seen two of them and read a lot of stories about in on Chimera's BB) still have issues with it.
As far as I know, this fix has not landed on the Chimera branch yet.
Brian, so should this bug reopened because this fix has not landed on the Chimera branch yet.
I landed the fix on the Chimera branch this morning. It will be in the next build.
Sorry if this is off topic, but has anyone got LiveConnect working again, even when using the method discussed in comment #44? I removed the netscape.* packages from ui.jar, but it still didn't work. I get null back when I call JSObject.getWindow(applet) in an applet. But the LiveConnect testcases at http://www.mozilla.org/quality/browser/front-end/testcases/oji/liveconnecttest.html works. LiveConnect is a big deal for a lot of applications, especially for financial and enterprise apps. Is there a workaround that we can use for now? What can be done by Mozilla people and what needs to be fixed by Apple engineers? Mozilla had an edge over IE on the Mac because of the LiveConnect support, so it would be great to get it to work again.
*** Bug 164630 has been marked as a duplicate of this bug. ***
wrt chimera, 2002.08.26.05 build: no longer crash loading about:plugins, or a page with java applets. (o'course the java applet itself didn't load, but that's prolly a different issue. ;) tested on Jaguar, and 10.1.5 to be sure.
*** Bug 164560 has been marked as a duplicate of this bug. ***
*** Bug 164622 has been marked as a duplicate of this bug. ***
Dunno if this is relevant, but 2002-08-22-11 showed the same symptoms, i.e. crashed on every Java applet, although in MOX 10.1.5. It doesn't crash in 2002-08-26-08.
removed fixed1.0.1 from keywords
Keywords: fixed1.0.1
*** Bug 164828 has been marked as a duplicate of this bug. ***
*** Bug 164927 has been marked as a duplicate of this bug. ***
as of Mozilla 1.1 -- i.e., Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.1) Gecko/20020826 -- bug 164622 still crashes Mozilla. now, though, it seems to happen at the original time of applet load rather than a subsidiary page after applet load. i.e., the crash occurs on Jaguar as soon as i visit http://www.wlac.com/ when did the fix land? was it in time for 1.1?
so should i re-open bug 164622, or has every dupe of this been resolved as of the post-20020826 trunk nightlies?
*** Bug 162872 has been marked as a duplicate of this bug. ***
This patch was not landed on the 1.1 branch, and therefore was not fixed in Mozilla 1.1. Eventually we will get a new plugin posted on mozilla.org. For now, I'd download a daily build and pull the plugin out of it.
*** Bug 165909 has been marked as a duplicate of this bug. ***
*** Bug 165900 has been marked as a duplicate of this bug. ***
*** Bug 165961 has been marked as a duplicate of this bug. ***
*** Bug 164768 has been marked as a duplicate of this bug. ***
*** Bug 166095 has been marked as a duplicate of this bug. ***
*** Bug 166113 has been marked as a duplicate of this bug. ***
*** Bug 166195 has been marked as a duplicate of this bug. ***
*** Bug 159884 has been marked as a duplicate of this bug. ***
*** Bug 165466 has been marked as a duplicate of this bug. ***
*** Bug 165513 has been marked as a duplicate of this bug. ***
*** Bug 166178 has been marked as a duplicate of this bug. ***
*** Bug 166378 has been marked as a duplicate of this bug. ***
*** Bug 166383 has been marked as a duplicate of this bug. ***
*** Bug 166432 has been marked as a duplicate of this bug. ***
*** Bug 166444 has been marked as a duplicate of this bug. ***
*** Bug 166512 has been marked as a duplicate of this bug. ***
*** Bug 165819 has been marked as a duplicate of this bug. ***
*** Bug 166908 has been marked as a duplicate of this bug. ***
*** Bug 167016 has been marked as a duplicate of this bug. ***
*** Bug 167092 has been marked as a duplicate of this bug. ***
*** Bug 165187 has been marked as a duplicate of this bug. ***
*** Bug 167352 has been marked as a duplicate of this bug. ***
*** Bug 167349 has been marked as a duplicate of this bug. ***
*** Bug 167402 has been marked as a duplicate of this bug. ***
*** Bug 167436 has been marked as a duplicate of this bug. ***
*** Bug 167465 has been marked as a duplicate of this bug. ***
*** Bug 167500 has been marked as a duplicate of this bug. ***
*** Bug 165104 has been marked as a duplicate of this bug. ***
*** Bug 167721 has been marked as a duplicate of this bug. ***
Reopening so this shows up on duplicates.cgi.
Status: VERIFIED → REOPENED
Resolution: FIXED → ---
Reresolving Fixed. Please don't re-mark Verified until 1.2 ships with this fix in it.
Status: REOPENED → RESOLVED
Closed: 23 years ago → 22 years ago
Resolution: --- → FIXED
*** Bug 167888 has been marked as a duplicate of this bug. ***
*** Bug 166991 has been marked as a duplicate of this bug. ***
*** Bug 167945 has been marked as a duplicate of this bug. ***
*** Bug 168272 has been marked as a duplicate of this bug. ***
*** Bug 168482 has been marked as a duplicate of this bug. ***
*** Bug 167888 has been marked as a duplicate of this bug. ***
*** Bug 168701 has been marked as a duplicate of this bug. ***
*** Bug 168856 has been marked as a duplicate of this bug. ***
This is with the new testcase site of <http://www.odcmp.com/> as the original testcase site now gives a 404 error alert
Bug # 154699 should be reopened as new bugs are being marked as duplicates of of it daily, and because I can now demonstrate that at least one apparent regression has occurred. Bug # 154699 now affects at least some variants of OS 10.x and also Mac OS 9.2.2 Using <http://www.odcmp.com/> as a testcase, as the original testcase site now gives a 404 error alert. Go to <http://www.odcmp.com/, and once the progress line at the bottom your page says "Applet Loaded', then click upon the top option, "About Us", of the left hand column of options. What should happen then is that you get taken to a page that tells you about the Civilian Marksmanship Program. [BTW this is the testcase site for Bug # 168872]. I cannot comment on what happens under OS 10.x as I do not have it installed. What happens on Mac OS 9.2.2 depends on which build of Mozilla you use, and also on how recent is your copy of the plug-in file 'MRJPlugin v.1.0b5'as the older one installed by "Mozilla/5.0 (Macintosh; U; PPC; en-US; rv:1.1) Gecko/20020810 Build ID 2002081011" has a datafork some 2,368 bytes larger [1%] than yesterday's variant of 'MRJPlugin v.1.0b5' installed by "Mozilla/5.0 (Macintosh; U; PPC; en-US; rv:1.2a) Gecko/20020915 Build ID 2002091503". If you do this testcase with Build ID 2002081011 and with the older 'MRJPlugin v.1.0b5, then you are taken to the page 'About Us' without incident. If you do this testcase with Build ID 2002081011 and with the newer 'MRJPlugin v.1.0b5, then Mozilla will crash. If you do this testcase with Build ID 2002091503 and with *either* variant of 'MRJPlugin v.1.0b5', then Mozilla will crash. See attached MacsBug stdlog file , attachment (id=99401), for the crash with the newer variant of 'MRJPlugin v.1.0b5'. From these 4 tests I conclude that there is something the matter with the newer smaller variant of the 'MRJPlugin v.1.0b5', and that also there is something the matter with recent builds of Mozilla such as Build ID 2002091503, and that both these apparent regressions occurred at some date after Aug. 10th 2002. The 2 other plug-in files installed by Mozilla also differ between Build ID 200208101 and Build ID 2002091503. "Default Plug-in" differs in its STR resource ID # 128 as different URLs are given for getting plug-ins from Netscape, and which both variants have the same size datafork the utility "ResCompare" reports their dataforks as differing. "MRJPlugin.jar" is the same size in both builds but the utility 'FileCompare' reports that their resource forks differ although 'ResCompare' reports these as identical. It would help if all files were given accurate version information.
*** Bug 168872 has been marked as a duplicate of this bug. ***
*** Bug 169464 has been marked as a duplicate of this bug. ***
*** Bug 169723 has been marked as a duplicate of this bug. ***
*** Bug 168871 has been marked as a duplicate of this bug. ***
*** Bug 169916 has been marked as a duplicate of this bug. ***
wighta@nmia.com, re: comment 106, this bug was about a specific problem with Mozilla and MRJ on Mac OS X 10.2. Problems with Java and Mozilla on Mac OS 9 have nothing to do with this bug and should be filed separately.
Blocks: majorbugs
*** Bug 170483 has been marked as a duplicate of this bug. ***
*** Bug 170141 has been marked as a duplicate of this bug. ***
*** Bug 171379 has been marked as a duplicate of this bug. ***
As of the daily build on September 27th, the Java applets in Jaguar still cause a browser crash once you move to another page. The problem occurs on most Java applets. Shouldn't this bug be reopened unless it was fixed in the past few hours?
*** Bug 171562 has been marked as a duplicate of this bug. ***
This bug has to do with the total inability to load java applets under 10.2. The bug you are describing has to do with unloading the applet, not loading it. If you have a reproducible case please file a new bug.
*** Bug 171732 has been marked as a duplicate of this bug. ***
*** Bug 166402 has been marked as a duplicate of this bug. ***
*** Bug 167304 has been marked as a duplicate of this bug. ***
*** Bug 169288 has been marked as a duplicate of this bug. ***
Alias: OSX_Java
*** Bug 171912 has been marked as a duplicate of this bug. ***
*** Bug 167690 has been marked as a duplicate of this bug. ***
Comment on attachment 99401 [details] binhexed MacsBug stdlog file for JavaScript crash under Mac OS 9.2.2 Not applicable.
Attachment #99401 - Attachment is obsolete: true
*** Bug 171847 has been marked as a duplicate of this bug. ***
*** Bug 172796 has been marked as a duplicate of this bug. ***
*** Bug 173965 has been marked as a duplicate of this bug. ***
Chris Petersen is a new QA contact for oji component. His email is: petersen@netscape.com
Assignee: beard → petersen
Setting qa contact to petersen@netscape.com
QA Contact: pmac → petersen
fixing small error for pmac@netscape.com (filter with : SPAMMAILSUCKS)
Assignee: petersen → joe.chou
*** Bug 165195 has been marked as a duplicate of this bug. ***
*** Bug 165447 has been marked as a duplicate of this bug. ***
*** Bug 175849 has been marked as a duplicate of this bug. ***
*** Bug 175739 has been marked as a duplicate of this bug. ***
*** Bug 176240 has been marked as a duplicate of this bug. ***
*** Bug 176487 has been marked as a duplicate of this bug. ***
*** Bug 176491 has been marked as a duplicate of this bug. ***
*** Bug 165018 has been marked as a duplicate of this bug. ***
*** Bug 180590 has been marked as a duplicate of this bug. ***
*** Bug 181758 has been marked as a duplicate of this bug. ***
*** Bug 181869 has been marked as a duplicate of this bug. ***
*** Bug 178047 has been marked as a duplicate of this bug. ***
Marking Verified per comment 96 as Mozilla 1.2 has shipped.
Status: RESOLVED → VERIFIED
*** Bug 179421 has been marked as a duplicate of this bug. ***
No longer blocks: majorbugs
Product: Core → Core Graveyard
Crash Signature: [@ JVM_RegisterSignal]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: