Problem when starting firefox-bin on 2.6.34.1

Discuss usability issues, general maintenance, and general support issues for a grsecurity-enabled system.

Re: Problem when starting firefox-bin on 2.6.34.1

Postby specs » Mon Aug 16, 2010 2:05 pm

I suggest you start a new thread with your problem since the problems posted here seem to be fixed.

Brad and the Pax-team will need some images, configurations and specific logs and/or error messages to produce a fix for your problem.
See http://en.wikibooks.org/wiki/Grsecurity#Reporting_Bugs
specs
 
Posts: 190
Joined: Sun Mar 26, 2006 7:00 am

Re: Problem when starting firefox-bin on 2.6.34.1

Postby PaX Team » Thu Aug 19, 2010 2:18 am

linkfanel wrote:
specs wrote:The traces in the logs (and the garbage on all terminals) are gone with the latest patch (grsecurity-2.2.0-2.6.34.3-201008112233)

Well it doesn't work for me, the problem is still the same with the latest patch :(
indeed, this bug wasn't fixed yet (and i found a related one since). since i'm a tad bit busy these days please be patient ;).
PaX Team
 
Posts: 2310
Joined: Mon Mar 18, 2002 4:35 pm

Re: Problem when starting firefox-bin on 2.6.34.1

Postby PaX Team » Mon Sep 13, 2010 6:28 am

PaX Team wrote:indeed, this bug wasn't fixed yet (and i found a related one since). since i'm a tad bit busy these days please be patient ;).
the latest patch should have the fix, can you guys confirm it?
PaX Team
 
Posts: 2310
Joined: Mon Mar 18, 2002 4:35 pm

Re: Problem when starting firefox-bin on 2.6.34.1

Postby specs » Mon Sep 13, 2010 8:56 am

What I feared from the 2.6.34.[4-6] patches is confirmed with the 2.6.34.2-pax-test16 patch.

With 2.6.34.2-pax-test13 I get a message on the terminal right after starting "firefox youtube.com".
With 2.6.34.2-pax-test16 I get no message on the terminal but I find in the logfiles (only with the test16-kernel):
Code: Select all
Sep 13 14:26:58 localhost syslogd 1.5.0#5: restart.
Sep 13 14:26:58 localhost kernel: klogd 1.5.0#5, log source = /proc/kmsg started.
Sep 13 14:26:58 localhost kernel: ver usbhid
Sep 13 14:26:58 localhost kernel: usbhid: USB HID core driver
Sep 13 14:26:58 localhost kernel: AUD_IN = 8
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x8 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x18 IDX=0x0
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0x23 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x23 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x23
Sep 13 14:26:58 localhost kernel: [Capture Source] NID=0x8, #SRC=1
Sep 13 14:26:58 localhost kernel:   [Mic] IDX=0x0
Sep 13 14:26:58 localhost kernel: AUD_OUT found 2
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0xc IDX=0x0
Sep 13 14:26:58 localhost /usr/sbin/cron[2026]: (CRON) INFO (pidfile fd = 3)
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0xc IDX=0x0
Sep 13 14:26:58 localhost /usr/sbin/cron[2027]: (CRON) STARTUP (fork ok)
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0xc
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x14 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x14
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x14
Sep 13 14:26:58 localhost kernel: AUD_OUT found 2
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0xc IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0xc IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0xc
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0x1b IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x1b IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x1b
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x1b
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0x8 IDX=0x0
Sep 13 14:26:58 localhost kernel: [Capture Switch] NID=0x8, DIR=IN, IDX=0x0
Sep 13 14:26:58 localhost kernel: [Capture Volume] NID=0x8, DIR=IN, IDX=0x0
Sep 13 14:26:58 localhost kernel: [Master Playback Volume] NID=0x2, DIR=OUT
Sep 13 14:26:58 localhost kernel: [Mic Playback Switch] NID=0xb, DIR=IN, IDX=0x0
Sep 13 14:26:58 localhost kernel: [Mic Playback Volume] NID=0xb, DIR=IN, IDX=0x0
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0xc IDX=0x1
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0xc IDX=0x1
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0xc
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x14 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x14
Sep 13 14:26:58 localhost kernel: [Front Mic Playback Switch] NID=0xb, DIR=IN, IDX=0x1
Sep 13 14:26:58 localhost kernel: [Front Mic Playback Volume] NID=0xb, DIR=IN, IDX=0x1
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0xc IDX=0x1
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0xc IDX=0x1
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0xc
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x14 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x14
Sep 13 14:26:58 localhost kernel: [Line Playback Switch] NID=0xb, DIR=IN, IDX=0x2
Sep 13 14:26:58 localhost kernel: [Line Playback Volume] NID=0xb, DIR=IN, IDX=0x2
Sep 13 14:26:58 localhost kernel: CONNECT: NID=0xc IDX=0x1
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0xc IDX=0x1
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0xc
Sep 13 14:26:58 localhost kernel: UNMUTE IN: NID=0x14 IDX=0x0
Sep 13 14:26:58 localhost kernel: UNMUTE OUT: NID=0x14
Sep 13 14:26:58 localhost kernel: uhci_hcd 0000:00:1d.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19


The playback of youtube movies is not possible because the sound is not initiated right. Playback of MP3 also does not work using mpg123.
The sound-problem is also visible sometimes when I start 2.6.34.4 with grsecurity (sometimes a reboot gives working sound) but I haven't seen the UNMUTE messages yet.
However I have not been able to use sound with 2.6.34.2-pax-test16 yet.

In short it seems one problem is fixed and a new one is shown. Unfortunately a less verbose problem.

Edit:
Above problem surfaces on a board with intel 945GC chipset, not on a netbook with 945GM.
specs
 
Posts: 190
Joined: Sun Mar 26, 2006 7:00 am

Re: Problem when starting firefox-bin on 2.6.34.1

Postby PaX Team » Mon Sep 13, 2010 11:08 am

specs wrote:In short it seems one problem is fixed and a new one is shown. Unfortunately a less verbose problem.

Edit:
Above problem surfaces on a board with intel 945GC chipset, not on a netbook with 945GM.
can you confirm that vanilla linux works fine with your sound device?
PaX Team
 
Posts: 2310
Joined: Mon Mar 18, 2002 4:35 pm

Re: Problem when starting firefox-bin on 2.6.34.1

Postby specs » Mon Sep 13, 2010 3:13 pm

It seems to be a nasty bug. It does not show itself each time.

The first time starting vanilla (2.6.34.2) sound was fine, second time it did not work.
Also with the latest grsecurity patch it did not work the first time, but works after a reboot.

The UNMUTE messages seem to be specific for 2.6.34.2 (not only for pax/grsec).
Another message seems to be older than the bug (hda-intel: IRQ timing workaround is activated for card #0. Suggest a bigger bdl_pos_adj).
I saw it often in combination, but not enough times to draw conclusions (statistics).

At least it does not only show itself with pax/grsecurity.

I can't find the original dumps or remains in pax-test16 or the newer grsecurity patches.
specs
 
Posts: 190
Joined: Sun Mar 26, 2006 7:00 am

Re: Problem when starting firefox-bin on 2.6.34.1

Postby linkfanel » Wed Sep 15, 2010 2:57 am

PaX Team wrote:the latest patch should have the fix, can you guys confirm it?

Yes, it disappeared with grsecurity-2.2.0-2.6.34.6-201009130754.patch. I also had a problem with my computer randomly freezing while browsing the internet, but so far so good, so it was probably related and is fixed now.
linkfanel
 
Posts: 39
Joined: Fri Jul 14, 2006 8:26 pm

Previous

Return to grsecurity support