[CLOSED] Brackets crash because thread cannot read from virtual address

[core-ai-bot]

Issue by simonhoss Thursday Nov 13, 2014 at 10:59 GMT Originally opened as https://github.com/adobe/brackets/issues/9920

---

Hi,

I develop a plugin which uses worker and IndexedDB. I have the problem that brackets crashes after a random time (sometimes 10 Minutes, sometime 4 hours), but always after typing, because the plugin reacts on user input and parses the text. I'm not able to reproduce the crash because of the randomness.

The exception is: The thread tried to read from or write to a virtual address for which it does not have the appropriate access.

Im working with Brackets 1.0, Windows 7 x64

I tried to start Brackets as Administrator, I also switched from IndexedDB to WebSQL but with no result.

Here is my dump: https://www.dropbox.com/s/rye1pwewktlryop/Brackets.exe.9604.dmp?dl=0

Any idea?

Greets

[core-ai-bot]

Comment by peterflynn Thursday Nov 13, 2014 at 18:14 GMT

---

Hmm, at a glance this looks like a V8 JIT bug that crashed one of the worker threads:

ExceptionAddress: 72656854
   ExceptionCode: c0000005 (Access violation)
  ExceptionFlags: 00000000
NumberParameters: 2
   Parameter[0]: 00000008
   Parameter[1]: 72656854
Attempt to execute non-executable address 72656854

...

FAULTING_THREAD:  000032bc

...

LAST_CONTROL_TRANSFER:  from 6004ebb7 to 72656854

STACK_TEXT:  
WARNING: Frame IP not in any known module. Following frames may be wrong.
06a7f818 6004ebb7 05f8bdc0 0b702540 06a7f858 0x72656854
06a7f828 5f5e8e7d 0b8958b0 05fdc560 0b702540 libcef!cef_time_delta+0x14921a7
06a7f858 5f89afbf 04b99f0c 04b99f0c 126e978d libcef!cef_time_delta+0xa2c46d
06a7f878 5f89b018 5f641f88 05f8bdc0 04b99f0c libcef!cef_time_delta+0xcde5af
06a7f894 5f64229e 05f8bdc0 00000001 00000000 libcef!cef_time_delta+0xcde608
06a7f8b8 5f708ebc 05f8bdc0 5f7096ac 00000000 libcef!cef_time_delta+0xa8588e
06a7f93c 5f84a211 04b99f00 05ee36c0 00000000 libcef!cef_time_delta+0xb4c4ac
06a7f958 5f845a76 04b99f00 00000000 00000000 libcef!cef_time_delta+0xc8d801
06a7f96c 6015a519 05ea8d68 2ff16c1c 00000000 libcef!cef_time_delta+0xc89066
06a7f9a4 6015a5a3 00000000 06a7f9bc 7559338a libcef!cef_time_delta+0x159db09
06a7f9b0 7559338a 05f3dd40 06a7f9fc 77749f72 libcef!cef_time_delta+0x159db93
06a7f9bc 77749f72 05f3dd40 71698d80 00000000 kernel32!BaseThreadInitThunk+0xe
06a7f9fc 77749f45 6015a53f 05f3dd40 00000000 ntdll!__RtlUserThreadStart+0x70
06a7fa14 00000000 6015a53f 05f3dd40 00000000 ntdll!_RtlUserThreadStart+0x1b

(the thread ID points to the 11th thread in the list, so it looks like it's a worker thread)

[core-ai-bot]

Comment by peterflynn Thursday Nov 13, 2014 at 18:18 GMT

---

Currently we use a pretty old version of Chromium to run Brackets -- equivalent to Chrome 29 -- so it's very possible there were V8 bugs back then that have been fixed more recently. The best thing might be to wait until we have a working branch with a newer Chromium, and see if the bug is gone there.

We're hoping to start migrating to a newer Chromium very soon, but if you're impatient, you could also search the Chromium bug database for issues that look similar to this, and see if/when they've been fixed.

[core-ai-bot]

Comment by simonhoss Thursday Nov 13, 2014 at 19:08 GMT

---

Thank you for your feedback! I will wait then.

[core-ai-bot]

Comment by peterflynn Monday Nov 24, 2014 at 21:17 GMT

---

@JeffryBooher Why is there a tracking label on this? I don't see a link to any upstream bug that we're waiting on. Removing label for now.

The 'cef' label should remind us to re-check this when we integrate the newer CEF, though.

[core-ai-bot]

Comment by simonhoss Wednesday Feb 25, 2015 at 20:33 GMT

---

No problems anymore with the new cef.

[core-ai-bot]

Comment by peterflynn Wednesday Feb 25, 2015 at 21:59 GMT

---

@simonhoss Awesome, thanks for checking back in!