Bug 1296117

Summary: [abrt] firefox: mozalloc_abort(char const*)(): plugin-container killed by SIGSEGV
Product: Red Hat Enterprise Linux 7 Reporter: Alexander Todorov <atodorov>
Component: firefoxAssignee: Martin Stransky <stransky>
Status: CLOSED WORKSFORME QA Contact: Desktop QE <desktop-qa-list>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 7.2CC: atodorov, mcsontos, nobody
Target Milestone: rc   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
Whiteboard: abrt_hash:440a5ab2937e0efbf5d274adbd67ab6556a8d02a
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-01-17 09:49:12 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
File: backtrace
none
File: cgroup
none
File: core_backtrace
none
File: dso_list
none
File: environ
none
File: exploitable
none
File: limits
none
File: machineid
none
File: maps
none
File: open_fds
none
File: proc_pid_status
none
File: var_log_messages
none
File: sosreport.tar.xz none

Description Alexander Todorov 2016-01-06 11:20:16 UTC
Version-Release number of selected component:
firefox-38.5.0-3.el7_2

Additional info:
reporter:       libreport-2.1.11
backtrace_rating: 4
cmdline:        /usr/lib64/firefox/plugin-container /usr/lib64/flash-plugin/libflashplayer.so -greomni /usr/lib64/firefox/omni.ja -appomni /usr/lib64/firefox/browser/omni.ja -appdir /usr/lib64/firefox/browser 5035 plugin
crash_function: mozalloc_abort(char const*)
executable:     /usr/lib64/firefox/plugin-container
global_pid:     5205
kernel:         3.10.0-327.el7.x86_64
runlevel:       N 5
type:           CCpp
uid:            1000

Truncated backtrace:
Thread no. 1 (9 frames)
 #0 mozalloc_abort(char const*) at /usr/lib64/firefox/libmozalloc.so
 #1 NS_DebugBreak at /usr/lib64/firefox/libxul.so
 #2 mozilla::ipc::MessageChannel::OnChannelErrorFromLink() at /usr/lib64/firefox/libxul.so
 #3 mozilla::ipc::ProcessLink::OnChannelError() at /usr/lib64/firefox/libxul.so
 #4 event_base_loop at /usr/lib64/firefox/libxul.so
 #5 base::MessagePumpLibevent::Run(base::MessagePump::Delegate*) at /usr/lib64/firefox/libxul.so
 #6 MessageLoop::Run() at /usr/lib64/firefox/libxul.so
 #7 base::Thread::ThreadMain() at /usr/lib64/firefox/libxul.so
 #8 ThreadFunc(void*) at /usr/lib64/firefox/libxul.so

Potential duplicate: bug 1168744

Comment 1 Alexander Todorov 2016-01-06 11:20:20 UTC
Created attachment 1112106 [details]
File: backtrace

Comment 2 Alexander Todorov 2016-01-06 11:20:21 UTC
Created attachment 1112107 [details]
File: cgroup

Comment 3 Alexander Todorov 2016-01-06 11:20:23 UTC
Created attachment 1112108 [details]
File: core_backtrace

Comment 4 Alexander Todorov 2016-01-06 11:20:25 UTC
Created attachment 1112109 [details]
File: dso_list

Comment 5 Alexander Todorov 2016-01-06 11:20:26 UTC
Created attachment 1112110 [details]
File: environ

Comment 6 Alexander Todorov 2016-01-06 11:20:28 UTC
Created attachment 1112111 [details]
File: exploitable

Comment 7 Alexander Todorov 2016-01-06 11:20:29 UTC
Created attachment 1112112 [details]
File: limits

Comment 8 Alexander Todorov 2016-01-06 11:20:31 UTC
Created attachment 1112113 [details]
File: machineid

Comment 9 Alexander Todorov 2016-01-06 11:20:33 UTC
Created attachment 1112114 [details]
File: maps

Comment 10 Alexander Todorov 2016-01-06 11:20:35 UTC
Created attachment 1112115 [details]
File: open_fds

Comment 11 Alexander Todorov 2016-01-06 11:20:36 UTC
Created attachment 1112116 [details]
File: proc_pid_status

Comment 12 Alexander Todorov 2016-01-06 11:20:37 UTC
Created attachment 1112117 [details]
File: var_log_messages

Comment 13 Alexander Todorov 2016-01-06 11:21:34 UTC
Created attachment 1112118 [details]
File: sosreport.tar.xz

Comment 15 Martin Stransky 2016-01-07 10:59:51 UTC
Is there any reproduction steps?

Comment 16 Alexander Todorov 2016-01-07 12:08:20 UTC
Nope, Firefox just crashed randomly while I was browsing. I don't know what caused it.

Comment 17 Marian Csontos 2016-01-27 17:31:01 UTC
Any chance it is related to video/media?

|mcsontos@deadbeef|17:20|~> firefox --no-remote --new-instance --ProfileManager
console.error: 
  [CustomizableUI]
  Custom widget with id loop-button does not return a valid node
console.error: 
  [CustomizableUI]
  Custom widget with id loop-button does not return a valid node
SystemMessageCache: init

Now soon after visiting pages usually with video (like https://www.elastic.co/products/kibana) I get an instant segfault:

DtsCreateShMem:unable to get shmid :13
DtsCreateShMem:unable to get shmid :13
Segmentation fault (core dumped)

###!!! [Child][MessageChannel] Error: Channel error: cannot send/recv


###!!! [Child][MessageChannel] Error: Channel error: cannot send/recv


###!!! [Child][MessageChannel] Error: Channel error: cannot send/recv


###!!! [Child][MessageChannel] Error: Channel error: cannot send/recv


###!!! [Child][MessageChannel] Error: Channel error: cannot send/recv

Comment 18 Marian Csontos 2016-01-28 13:09:52 UTC
I tried disabling autoplay (set media.autoplay.enabled in about:config to false) and NoScript but still got a failure on the above page.

Thought possible workaround could be installing "Stop Tube HTML5" add on, as installing the add-on I could browse for a little longer, but the crash still occurred, subjectively later.

I am on Lenovo W541 - this has 2 video cards - one built-in Intel and NVidia.
I am wondering if FF is opening a media stream to video device when there is one whether it is playing or not, and it gets confused...

I am on RHEL 7.2 with recent kernel:

Linux deadbeef.usersys.redhat.com 3.10.0-327.4.4.el7.x86_64 #1 SMP Thu Dec 17 15:51:24 EST 2015 x86_64 x86_64 x86_64 GNU/Linux

And using the Xorg drivers:

xorg-x11-drv-nouveau-1.0.11-2.el7.x86_64
xorg-x11-drv-intel-2.99.917-8.20150615.el7.x86_64

Alexander, what's your HW?

Do you have SElinux enabled? I do and found the following in the logs:

----
time->Thu Jan 28 11:40:48 2016
type=SYSCALL msg=audit(1453977648.055:29480): arch=c000003e syscall=2 success=no exit=-13 a0=2038580 a1=2 a2=0 a3=7ffdf0be1c30 items=0 ppid=4920 pid=5084 auid=13940 uid=13940 gid=13940 euid=13940 suid=13940 fsuid=13940 egid=13940 sgid=13940 fsgid=13940 tty=(none) ses=1 comm="gst-plugin-scan" exe="/usr/libexec/gstreamer-1.0/gst-plugin-scanner" subj=unconfined_u:unconfined_r:sandbox_web_t:s0:c40,c330 key=(null)
type=AVC msg=audit(1453977648.055:29480): avc:  denied  { read write } for  pid=5084 comm="gst-plugin-scan" name="video0" dev="devtmpfs" ino=16863 scontext=unconfined_u:unconfined_r:sandbox_web_t:s0:c40,c330 tcontext=system_u:object_r:v4l_device_t:s0 tclass=chr_file

This was FF used in SElinux sandbox.
This does not map 1:1 to all the crashes I have seen, but could this be the culprit?

selinux-policy-targeted-3.13.1-60.el7.noarch

I wonder whether the "exit=-13" here:

    type=SYSCALL msg=audit(1453977648.055:29480): arch=c000003e syscall=2 success=no exit=-13 

is the same 13 as seen on console:

    DtsCreateShMem:unable to get shmid :13

May be we are up to something...

/me is rebooting to a newer kernel...

Comment 19 Alexander Todorov 2016-01-28 13:58:26 UTC
(In reply to Marian Csontos from comment #18)
> 
> Alexander, what's your HW?
> 
> Do you have SElinux enabled? I do and found the following in the logs:
> 


Hi Marian,
I'm on a MacBook Air 13", early 2015 model. I have latest RHEL 7.2 updated. 

This was a random FF crash so I have no idea what really caused it and can't reproduce. Opening the above website works for me and I can play the vimeo video. Also IIRC I've opted for HTML5 on YouTube as well. I have SELinux in Enforcing mode. I believe the Xorg driver is Intel. 

Here's what I have installed and some HW specs:

[atodorov@aero ~]$ uname -r
3.10.0-327.4.5.el7.x86_64
firefox-38.6.0-1.el7_2.x86_64
xorg-x11-drv-intel-2.99.917-8.20150615.el7.x86_64

Detailed hardware listings can be found here:
http://atodorov.org/blog/2015/04/26/installing-red-hat-enterprise-linux-7-on-macbook-air-2015/


I've installed all -debuginfo packages for FF and I'll try to resubmit the backtrace with more details if I still keep it locally.

Comment 20 Marian Csontos 2016-01-28 15:28:41 UTC
For now after reboot not reproducible.

I have two monitors, and I am frequently docking/undocking and suspending/resuming the notebook, so let's see how long it will last...

Comment 21 Martin Stransky 2016-02-26 14:33:39 UTC
That should be fixed by update to ESR45 which contains a fix for that. Let's wait until ESR45 is out and then retest.

Comment 22 Martin Stransky 2018-01-17 09:49:12 UTC
Should be fixed now.