Closed Bug 528138 Opened 10 years ago Closed 7 years ago

Random crash in JavaEmbeddingPlugin during test_HTMLDocument07.html


(Plugins Graveyard :: Java (Java Embedding Plugin), defect, critical)

Not set


(Not tracked)



(Reporter: kinetik, Assigned: smichaud)



(Keywords: crash, intermittent-failure)

50520 INFO Running /tests/dom/tests/mochitest/dom-level2-html/test_HTMLDocument07.html...
Wed Nov 11 19:51:50 PST 2009 JEP creating applet org.w3c.domts.DOMTSApplet (http://localhost:8888/tests/dom/tests/mochitest/dom-level2-html/files/applets/)
2009-11-11 19:51:51.216 firefox-bin[788:1481b] Invalid memory access of location c0000023 eip=919a06e8
TEST-UNEXPECTED-FAIL | | Exited with code 1 during test run

Top of stack dump:
Crash address: 0xffffffffc0000023

Thread 0 (crashed)
 0  libobjc.A.dylib + 0x146e8
    eip = 0x919a06e8   esp = 0xbfffdaf8   ebp = 0xbfffdb18   ebx = 0x079a58d4
    esi = 0x10281590   edi = 0x100a8190   eax = 0x0fc90b70   ecx = 0x07b030e6
    edx = 0xc0000003   efl = 0x00210202
    Found by: given as instruction pointer in context
 1  JavaEmbeddingPlugin + 0x4b3f
    eip = 0x06593b40   esp = 0xbfffdb20   ebp = 0xbfffdb38
    Found by: previous frame's frame pointer
 2  Foundation + 0xaada
    eip = 0x9673cadb   esp = 0xbfffdb40   ebp = 0xbfffdc68
    Found by: previous frame's frame pointer
 3  Foundation + 0x5e027
    eip = 0x96790028   esp = 0xbfffdc70   ebp = 0xbfffdcb8
    Found by: previous frame's frame pointer
 4  JavaEmbeddingPlugin + 0x8227
    eip = 0x06597228   esp = 0xbfffdcc0   ebp = 0xbfffdd08
    Found by: previous frame's frame pointer

Full log:
Whiteboard: [orange]
Mochitest run before and after this crash were green, so this looks random.
The crash log doesn't give me anywhere near enough information to work
with.  So we need to start looking for patterns.

1) How often do these crashes happen?

2) In which tests do they happen?

   There are a lot of other tests besides test_HTMLDocument07.html
   that load Java applets.

3) On which branches do they happen?

   test_HTMLDocument07.html appears to be exactly the same on the
   1.9.1 branch as it is on the 1.9.2 branch.  This is probably also
   true of many of the other tests that load Java applets.
This bug might be easier to figure out if the Mochitest crash logs
contained symbol information (including for the system calls, like
those from the Foundation framework).

Ted, do you know if this is something we can do?
It's a bit of a pain because we're generating the stack trace using Breakpad, so we'd have to have the system symbols available in Breakpad format. I think it's doable, I've filed bug 528231 with a proposal.
Severity: normal → critical
Keywords: crash
Component: Java Embedding Plugin → Java (Java Embedding Plugin)
Product: Core → Plugins
Version: 1.9.2 Branch → unspecified
Mass marking whiteboard:[orange] bugs WFM (to clean up TBPL bug suggestions) that:
* Haven't changed in > 6months
* Whose whiteboard contains none of the strings: {disabled,marked,random,fuzzy,todo,fails,failing,annotated,leave open,time-bomb}
* Passed a (quick) manual inspection of bug summary/whiteboard to ensure they weren't a false positive.

I've also gone through and searched for cases where the whiteboard wasn't labelled correctly after test disabling, by using attachment description & basic comment searches. However if the test for which this bug was about has in fact been disabled/annotated/..., please accept my apologies & reopen/mark the whiteboard appropriately so this doesn't get re-closed in the future (and please ping me via IRC or email so I can try to tweak the saved searches to avoid more edge cases).

Sorry for the spam! Filter on: #FFA500
Closed: 7 years ago
Resolution: --- → WORKSFORME
Whiteboard: [orange]
Product: Plugins → Plugins Graveyard
You need to log in before you can comment on or make changes to this bug.