PDA

View Full Version : Heads up: Nvidia possibly vulnerable



towsonu2003
October 17th, 2006, 12:25 AM
I'm directly copy pasting:


The NVIDIA Binary Graphics Driver for Linux is vulnerable to a
buffer overflow that allows an attacker to run arbitrary code as
root.

from http://download2.rapid7.com/r7-0025/

just wanted to make sure you know :)

PS. yes, I saw it on slashdot ;)

Note: this is being tracked as a bug at https://launchpad.net/distros/ubuntu/+source/linux-restricted-modules-2.6.17/+bug/46034
Note2: the open source nv driver is fine.
Note3: if nvidia has this, I can't imagine what fglrx has... ouch...

Animortis
October 17th, 2006, 05:57 AM
Rapid7 posted this flaw (http://download2.rapid7.com/r7-0025/) today, along with how to do it, and it appeared in OSnews.com (http://www.osnews.com/) as a security exploit.

It's an exploit that allows the takeover and root access of a system running nvidia drivers. It's an old exploit that nvidia had never gotten around to fixing.

So, how seriously should we be taking this exploit? I have switched to nv drivers for now, and while I prefer open-source stuff whenever possible, the performance of the proprietary drivers cannot be ignored. Should we wait for the fixed drivers to update in the repos (Could take weeks, months) or should we ignore this exploit since it's been out since 2004 and it's not caused any problems before? (Though posting the exploit should encourage faster fixing, it'll also encourage more exploitation.)

I miss my linux-based games that require 3d accelleration.

I am looking for non-FUD-based opinions. The site OSNews links, Kernel Trap, has discussion areas and they're full of trolls and flames. I am looking for real opinions here.

I am not the most technical user, but I know some things. Let's try to keep it in (complex) English, please.

kerry_s
October 17th, 2006, 06:46 AM
I use nvidia and i ain't worried. Also linux is not on the list->

PROBABLY VULNERABLE:
o NVIDIA Driver for FreeBSD
o NVIDIA Driver for Solaris
o Earlier versions

FreeBSD and Solaris is different than what ubuntu uses.

Zeddicus
October 17th, 2006, 06:51 AM
I use nvidia and i ain't worried. Also linux is not on the list->

PROBABLY VULNERABLE:
o NVIDIA Driver for FreeBSD
o NVIDIA Driver for Solaris
o Earlier versions


KNOWN VULNERABLE:
o NVIDIA Driver For Linux v8774
o NVIDIA Driver For Linux v8762

;)

kerry_s
October 17th, 2006, 06:58 AM
Da, i really gotta start reading instead of skimming through ](*,) or start just doing one thing at a time.LOL I've got duel monitors and i use them :mrgreen:

peter07
October 17th, 2006, 07:20 AM
I think switching to "nv" is good idea ( that means changing the xorg.conf file or I need to change something else too :confused: ?)

chaosgeisterchen
October 17th, 2006, 07:22 AM
Doesn't seem all too good what one is reading here...

but how come that nVidia drivers act as they seem to do?

kerry_s
October 17th, 2006, 07:26 AM
I don't know i took the test but didn't get no crash.

-> http://nvidia.com/content/license/location_0605.asp?url=';a='a';i=18;while(i--)a%2b=a;location=a;//

deepwave
October 17th, 2006, 07:38 AM
A buffer overflow security flaw for any software is a VERY serious issue. It doesn't take much to exploit, and compromise the system. All you need is a shellcode to overwrite the buffer, and poff! One remote shell running. And the shell will run the permissions of the compromised program.

In this case since the nvidia driver runs as root, so a shell run from this exploit would run with root's permissions.

woedend
October 17th, 2006, 08:31 AM
is the 9xxx series affected? its not listed, and what im using.

Zeddicus
October 17th, 2006, 09:14 AM
is the 9xxx series affected? its not listed, and what im using.

Supposedly, it's fixed in the beta 9xxx drivers.

JamesB
October 17th, 2006, 09:38 AM
This link seems to show that it has been fixed in the Beta version 9xxx.
Beware, there is a link on this page that could crash your X if you are using firefox, not so opera:

http://kerneltrap.org/node/7228#comment-200983

Here there is a suggested work around:

http://www.nvnews.net/vbulletin/showthread.php?t=78322

also however as most distros have X remote session disabled by default, then unless you are using it, you are safe.

asimon
October 17th, 2006, 10:23 AM
Supposedly, it's fixed in the beta 9xxx drivers.
But what about updated legacy drivers? There are still machines with older NVIDIA cards in use. How NVIDIA deals with this issue doesn't impress me. I am disappointed.

zenwhen
October 17th, 2006, 11:58 AM
I don't know i took the test but didn't get no crash.

-> http://nvidia.com/content/license/location_0605.asp?url=';a='a';i=18;while(i--)a%2b=a;location=a;//

I can confirm that this test crashed my X server when I used Firefox, and did not crash my X server after I updated to the beta driver. If we are going to keep the proprietary driver in the repos, the beta driver should be used.

The fact that this worked in Firefox and not Opera with this link does not mean that there is no way to exploit this in Opera.

jamespi
October 17th, 2006, 01:01 PM
Does anyone know if this (WARNING! SEE BELOW* BEFORE CLICKING THIS LINK: http://kerneltrap.org/node/7228) affects the Repo versions of the Nvidia drivers? I have Dapper, but I imagine this will need to be checked for other releases too.

Apparently Nvidia have released fixed drivers (http://www.nvidia.com/object/linux_display_ia32_1.0-9626.html). Can these be added to the Repo's then pushed out as a update, or will we have to manually install them?

*Someone added a comment to that article with a link to an exploit for a firefox bug. the comment reads:
"Here is nice special link the authors of the advisory have shown a few people who say this "this vuln is primarily local".
DONT CLICK IT.

bastiegast
October 17th, 2006, 01:25 PM
is the 9xxx series affected? its not listed, and what im using.

I use them too, When are these drivers gonna be out of beta? Cant seem to find it anywhere.

MKR.
October 17th, 2006, 01:45 PM
Why is the noscript plugin not included in a default Firefox install?

So handy. :confused:

aurelieng
October 17th, 2006, 03:02 PM
Is the nvidia beta driver available as a deb package for dapper ?

not-a-bot
October 17th, 2006, 03:09 PM
@MKR

You are aware, that NoScript wouldn't change anything at all? This is not a Javascript issue!

@all

This is a very serious flaw and I personally would hope that the 9xxx-series drivers, that have this vulnerability fixed, will be made available in the repositories (and included in Edgy) as soon as possible. Most known bugs in these beta-drivers are related to the new features that cannot be used with Xorg 7.0 (and therefor in the default Dapper setup) anyways.

For the legacy driver ... I have no idea ...

Polygon
October 17th, 2006, 03:11 PM
Why is the noscript plugin not included in a default Firefox install?

So handy. :confused:

noscript couldent prevent against that, as i believe the point is that if the drivers render the text in that url, which is a line of code supposedly creating a bufferoverflow, its exploiting the exploit that we are talking about in this topic

not-a-bot
October 17th, 2006, 03:17 PM
Is the nvidia beta driver available as a deb package for dapper ?

As far as I know ... they aren't any packages, at least not officially. And using the nvidia-installer or unofficial packages is a royal pain in the rectum because they need to be updated or reinstalled after every kernelupdate and may conflict with other closed-source-kernel-add-ons e.g. for wireless adapters.
I wish there would be officially maintained 9xxx-drivers for Ubuntu.

DoctorMO
October 17th, 2006, 03:24 PM
This problem shows us the problems with propritory drivers. people keep on harping on about how including propritory code isn't so bad.

But it _is_ bad because no one here no matter how good their skill can fix this issue.

We should write to nVidia to get them to open source their drivers so we can make use of it and make it safe. no way would this kind of bug exist in an open sourced version for so long.

dannyboy79
October 17th, 2006, 03:26 PM
can some1 tell me how to quickly identify which version of the nvidia driver I am running? I have a GeForce 6200. is that using the newest beta driver? is there a way to tell me by typing something into a terminal? cause when I go into my xorg.conf, all it says is driver: nvidia, I think. ( i am not at home?)

aurelieng
October 17th, 2006, 03:35 PM
@dannyboy79: type in a terminal :

grep "NVIDIA X Driver" /var/log/Xorg.0.log

carlgm
October 17th, 2006, 03:37 PM
We should write to nVidia to get them to open source their drivers so we can make use of it and make it safe. no way would this kind of bug exist in an open sourced version for so long.

Send the letter to yourself, first class stamp. It would be more useful.

I wish people would stop going on about the fact that the main nvidia drivers are closed source. There is an optional open source driver available, although still without 3D acceleration I believe.

Are you using the drivers from nvidia.com or apt-get?

evilghost
October 17th, 2006, 04:01 PM
Yes it affects the drivers in the Repo. It's only patched in the 1.0.9XXX series. You can mitigate the vulnerability by setting RenderAccel to false.

http://www.nvnews.net/vbulletin/showpost.php?p=1027960&postcount=2

Perfect Storm
October 17th, 2006, 04:01 PM
As far as I know ... they aren't any packages, at least not officially. And using the nvidia-installer or unofficial packages is a royal pain in the rectum because they need to be updated or reinstalled after every kernelupdate and may conflict with other closed-source-kernel-add-ons e.g. for wireless adapters.
I wish there would be officially maintained 9xxx-drivers for Ubuntu.

9xxx drivers are only for card specifically for Quadro Plex setups, ofcause if you have such card I envy you :mrgreen:

PriceChild
October 17th, 2006, 04:18 PM
9xxx drivers are only for card specifically for Quadro Plex setups, ofcause if you have such card I envy you :mrgreen:That's the official release... 9626

9625 betas can be found here for all: http://www.nzone.com/object/nzone_downloads_rel70betadriver.html

evilghost
October 17th, 2006, 04:19 PM
9xxx drivers are only for card specifically for Quadro Plex setups, ofcause if you have such card I envy you :mrgreen:

This is simply not true, I am running a Go FX5700 with 1.0.9626. As mentioned before disabling RenderAccel will mitigate this vulnerability or you can upgrade to the 1.0.9XXX series driver.

1.0.9626 Changelog:
* Adds support for NVIDIA Quadro Plex configurations"

1.0.9525 Changelog:
* Added initial support for GLX_EXT_texture_from_pixmap.
* Added new "Display Configuration" page in nvidia-settings.
* Improved workstation OpenGL performance in Xinerama.
* Added support for NVIDIA Quadro Plex.
* Added support for Quad SLI.
* Improved X driver error recovery.
* Improved workstation overlay performance.
* Added SMBus functionality to the Linux/i2c interface.
* Fixed DFP scaling support.
* Added support for OpenGL 2.1.
* Added new "TwinViewXineramaInfoOrder" X configuration option to control the order of display devices when in TwinView.

The 1.0.9626 driver is available here: ftp://download.nvidia.com/XFree86/Linux-x86/1.0-9626

The ReadMe located at ftp://download.nvidia.com/XFree86/Linux-x86/1.0-9626/README/README.txt clearly states the following cards are supported:



NVIDIA chip name Device PCI ID
---------------------------------- ----------------------------------
GeForce 6800 Ultra 0x0040
GeForce 6800 0x0041
GeForce 6800 XE 0x0043
GeForce 6800 XT 0x0044
GeForce 6800 GT 0x0045
GeForce 6800 GT 0x0046
GeForce 6800 GS 0x0047
GeForce 6800 XT 0x0048
Quadro FX 4000 0x004E
GeForce 7800 GTX 0x0090
GeForce 7800 GTX 0x0091
GeForce 7800 GT 0x0092
GeForce 7800 GS 0x0093
GeForce Go 7800 0x0098
GeForce Go 7800 GTX 0x0099
Quadro FX 4500 0x009D
GeForce 6800 GS 0x00C0
GeForce 6800 0x00C1
GeForce 6800 LE 0x00C2
GeForce 6800 XT 0x00C3
GeForce Go 6800 0x00C8
GeForce Go 6800 Ultra 0x00C9
Quadro FX Go1400 0x00CC
Quadro FX 3450/4000 SDI 0x00CD
Quadro FX 1400 0x00CE
GeForce 6800/GeForce 6800 Ultra 0x00F0
GeForce 6600/GeForce 6600 GT 0x00F1
GeForce 6600 0x00F2
GeForce 6200 0x00F3
GeForce 6600 LE 0x00F4
GeForce 7800 GS 0x00F5
GeForce 6800 GS 0x00F6
Quadro FX 3400/4400 0x00F8
GeForce 6800 Ultra 0x00F9
GeForce PCX 5750 0x00FA
GeForce PCX 5900 0x00FB
Quadro FX 330/GeForce PCX 5300 0x00FC
Quadro NVS 280 PCI-E/Quadro FX 330 0x00FD
Quadro FX 1300 0x00FE
GeForce PCX 4300 0x00FF
GeForce2 MX/MX 400 0x0110
GeForce2 MX 100/200 0x0111
GeForce2 Go 0x0112
Quadro2 MXR/EX/Go 0x0113
GeForce 6600 GT 0x0140
GeForce 6600 0x0141
GeForce 6600 LE 0x0142
GeForce 6600 VE 0x0143
GeForce Go 6600 0x0144
GeForce 6610 XL 0x0145
GeForce Go 6600 TE/6200 TE 0x0146
GeForce Go 6600 0x0148
GeForce Go 6600 GT 0x0149
Quadro NVS 440 0x014A
Quadro FX 550 0x014C
Quadro FX 540 0x014E
GeForce 6200 0x014F
GeForce 6500 0x0160
GeForce 6200 TurboCache(TM) 0x0161
GeForce Go 6200 0x0164
Quadro NVS 285 0x0165
GeForce Go 6400 0x0166
GeForce Go 6200 0x0167
GeForce Go 6400 0x0168
GeForce4 MX 460 0x0170
GeForce4 MX 440 0x0171
GeForce4 MX 420 0x0172
GeForce4 MX 440-SE 0x0173
GeForce4 440 Go 0x0174
GeForce4 420 Go 0x0175
GeForce4 420 Go 32M 0x0176
GeForce4 460 Go 0x0177
Quadro4 550 XGL 0x0178
GeForce4 440 Go 64M 0x0179
Quadro NVS 0x017A
Quadro4 500 GoGL 0x017C
GeForce4 410 Go 16M 0x017D
GeForce4 MX 440 with AGP8X 0x0181
GeForce4 MX 440SE with AGP8X 0x0182
GeForce4 MX 420 with AGP8X 0x0183
GeForce4 MX 4000 0x0185
Quadro4 580 XGL 0x0188
Quadro NVS with AGP8X 0x018A
Quadro4 380 XGL 0x018B
Quadro NVS 50 PCI 0x018C
GeForce2 Integrated GPU 0x01A0
GeForce 7300 LE 0x01D1
Quadro NVS 110M 0x01D7
GeForce Go 7300 0x01D7
GeForce Go 7400 0x01D8
Quadro NVS 110M 0x01DA
Quadro NVS 120M 0x01DB
Quadro FX 350M 0x01DC
Quadro FX 350 0x01DE
GeForce 7300 GS 0x01DF
GeForce4 MX Integrated GPU 0x01F0
GeForce3 0x0200
GeForce3 Ti 200 0x0201
GeForce3 Ti 500 0x0202
Quadro DCC 0x0203
GeForce 6800 0x0211
GeForce 6800 LE 0x0212
GeForce 6800 GT 0x0215
GeForce 6800 XT 0x0218
GeForce 6150 0x0240
GeForce 6150 LE 0x0241
GeForce 6100 0x0242
GeForce4 Ti 4600 0x0250
GeForce4 Ti 4400 0x0251
GeForce4 Ti 4200 0x0253
Quadro4 900 XGL 0x0258
Quadro4 750 XGL 0x0259
Quadro4 700 XGL 0x025B
GeForce4 Ti 4800 0x0280
GeForce4 Ti 4200 with AGP8X 0x0281
GeForce4 Ti 4800 SE 0x0282
GeForce4 4200 Go 0x0286
Quadro4 980 XGL 0x0288
Quadro4 780 XGL 0x0289
Quadro4 700 GoGL 0x028C
GeForce 7900 GTX 0x0290
GeForce 7900 GT 0x0291
GeForce Go 7900 GS 0x0298
GeForce Go 7900 GTX 0x0299
Quadro FX 2500M 0x029A
Quadro FX 1500M 0x029B
Quadro FX 5500 0x029C
Quadro FX 3500 0x029D
Quadro FX 1500 0x029E
Quadro FX 4500 X2 0x029F
GeForce 7600 GS 0x02E1
GeForce FX 5800 Ultra 0x0301
GeForce FX 5800 0x0302
Quadro FX 2000 0x0308
Quadro FX 1000 0x0309
GeForce FX 5600 Ultra 0x0311
GeForce FX 5600 0x0312
GeForce FX 5600XT 0x0314
GeForce FX Go5600 0x031A
GeForce FX Go5650 0x031B
Quadro FX Go700 0x031C
GeForce FX 5200 0x0320
GeForce FX 5200 Ultra 0x0321
GeForce FX 5200 0x0322
GeForce FX 5200LE 0x0323
GeForce FX Go5200 0x0324
GeForce FX Go5250 0x0325
GeForce FX 5500 0x0326
GeForce FX 5100 0x0327
GeForce FX Go5200 32M/64M 0x0328
Quadro NVS 280 PCI 0x032A
Quadro FX 500/600 PCI 0x032B
GeForce FX Go53xx 0x032C
GeForce FX Go5100 0x032D
GeForce FX 5900 Ultra 0x0330
GeForce FX 5900 0x0331
GeForce FX 5900XT 0x0332
GeForce FX 5950 Ultra 0x0333
GeForce FX 5900ZT 0x0334
Quadro FX 3000 0x0338
Quadro FX 700 0x033F
GeForce FX 5700 Ultra 0x0341
GeForce FX 5700 0x0342
GeForce FX 5700LE 0x0343
GeForce FX 5700VE 0x0344
GeForce FX Go5700 0x0347
GeForce FX Go5700 0x0348
Quadro FX Go1000 0x034C
Quadro FX 1100 0x034E
GeForce 7600 GT 0x0391
GeForce 7600 GS 0x0392
GeForce Go 7600 0x0398
Quadro FX 560 0x039E


Below are the legacy GPUs that are no longer supported in the unified driver.
These GPUs will continue to be maintained through the special legacy NVIDIA
GPU driver releases.


NVIDIA chip name Device PCI ID
---------------------------------- ----------------------------------
RIVA TNT 0x0020
RIVA TNT2/TNT2 Pro 0x0028
RIVA TNT2 Ultra 0x0029
Vanta/Vanta LT 0x002C
RIVA TNT2 Model 64/Model 64 Pro 0x002D
Aladdin TNT2 0x00A0
GeForce 256 0x0100
GeForce DDR 0x0101
Quadro 0x0103
GeForce2 GTS/GeForce2 Pro 0x0150
GeForce2 Ti 0x0151
GeForce2 Ultra 0x0152
Quadro2 Pro 0x0153

Perfect Storm
October 17th, 2006, 04:20 PM
Ah, okay. Then I misunderstood the release note.

evilghost
October 17th, 2006, 04:23 PM
Ah, okay. Then I misunderstood the release note.

No problem, I just wanted to ensure everyone fully understood the situation, there seems to be a fair amount of FUD around this issue. :)

PriceChild
October 17th, 2006, 04:27 PM
I wish there would be officially maintained 9xxx-drivers for Ubuntu.One could argue this is a serious security fix which should be added :P Doubt you'd win though ;)

OffHand
October 17th, 2006, 05:01 PM
I just tried the link on a up-to-date Edgy install with Firefox.
Nothing happened. I am not running the beta drives and direct rendering is enabled.

$ grep "NVIDIA X Driver" /var/log/Xorg.0.log
NVIDIA X Driver 1.0-8774 Tue Aug 1 20:56:50 PDT 2006

$ glxinfo | grep rendering
direct rendering: Yes

dannyboy79
October 17th, 2006, 07:11 PM
WOW, I am running the 8762 driver and my sudo lspci -v states:
0000:01:00.0 VGA compatible controller: nVidia Corporation GeForce 6200 (rev a1) (prog-if 00 [VGA])
Subsystem: XFX Pine Group Inc.: Unknown device 2144
Flags: bus master, VGA palette snoop, 66MHz, medium devsel, latency 248, IRQ 169
Memory at e8000000 (32-bit, non-prefetchable) [size=16M]
Memory at d0000000 (32-bit, prefetchable) [size=256M]
Memory at e9000000 (32-bit, non-prefetchable) [size=16M]
Expansion ROM at ea000000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
Capabilities: [44] AGP version 3.0
Could I run the latest and greatest like evilghost is since he has an older card than i do? I think i merely installed what was in Synaptic! :-(

evilghost
October 17th, 2006, 07:16 PM
WOW, I am running the 8762 driver and my sudo lspci -v states:
0000:01:00.0 VGA compatible controller: nVidia Corporation GeForce 6200 (rev a1) (prog-if 00 [VGA])
Subsystem: XFX Pine Group Inc.: Unknown device 2144
Flags: bus master, VGA palette snoop, 66MHz, medium devsel, latency 248, IRQ 169
Memory at e8000000 (32-bit, non-prefetchable) [size=16M]
Memory at d0000000 (32-bit, prefetchable) [size=256M]
Memory at e9000000 (32-bit, non-prefetchable) [size=16M]
Expansion ROM at ea000000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
Capabilities: [44] AGP version 3.0
Could I run the latest and greatest like evilghost is since he has an older card than i do? I think i merely installed what was in Synaptic! :-(

Yes, you can run 1.0.9626. I compiled mine from the nVidia package.

not-a-bot
October 17th, 2006, 08:50 PM
One could argue this is a serious security fix which should be added :P Doubt you'd win though ;)

Well, facts are that the currently used binary drivers in Dapper and Edgy have a very, very serious security flaw. Since they are not open source this cannot be easily be fixed.
There are only two possible options for Edgy before shipping it, if security is taken seriously by the developers:
* Remove the affected nvidia-package and ship it without the binary driver.
* Replace it with a driver from the 9xxx-series that does NOT have this security flaw.

I can see only one satisfactory option for Dapper:
* Update the affected package to the current 9xxx-drivers as soon as possible to close this gaping hole!


I understand why you are laughing about this issue, god beware that we get a feature update together with a security update. But currently there seems to be no good way around that.

I don't really see a big problem in issuing a security update to the 9xxx-drivers, they don't seem to be less stable than the currently used drivers in most cases and even bring optional functionality that users can benefit from if they want.

Colonel Kilkenny
October 17th, 2006, 09:09 PM
Can someone explain what code in browser url has to do with Nvidia? What is the mechanism behind that exploit? I understand that browsers with unsufficient security measures can execute that code but how can it crash X? Why doesn't it crash just the browser?

As you can see, I'm quite out of these sort of things...

Btw. I think Opera scans url's for exploits (mainly because SQL injection I think) and that's why it won't work at least in that form in Opera.

brt
October 17th, 2006, 10:29 PM
does anyone know when we can expect this driver in the repos?

dabdine
October 17th, 2006, 10:36 PM
FireFox doesn't properly sanitize (in other words, clip to a certain length) strings entered in its controls, or even text rendered in html input fields, etc. This is why you can use FireFox as an attack vector and not Konqueror (or Opera). One could argue this is an unnecessary risk that Mozilla could easily remove from the browser, but it is not a vulnerability in the browser itself.

-Derek

OffHand
October 17th, 2006, 10:40 PM
FireFox doesn't properly sanitize (in other words, clip to a certain length) strings entered in its controls, or even text rendered in html input fields, etc. This is why you can use FireFox as an attack vector and not Konqueror (or Opera). One could argue this is an unnecessary risk that Mozilla could easily remove from the browser, but it is not a vulnerability in the browser itself.

-Derek

So why didn't the test page do anything at all on my machine using Firefox?

dabdine
October 17th, 2006, 11:37 PM
Either:
* You are not using the 8774 or 8762 driver. Check nvidia-settings.
* You (for some reason) do not have glyph rendering acceleration enabled.

OffHand
October 18th, 2006, 12:13 AM
Either:
* You are not using the 8774 or 8762 driver. Check nvidia-settings.
* You (for some reason) do not have glyph rendering acceleration enabled.

$ grep "NVIDIA X Driver" /var/log/Xorg.0.log
NVIDIA X Driver 1.0-8774 Tue Aug 1 20:56:50 PDT 2006

$ glxinfo | grep rendering
direct rendering: Yes

What is glyph rendering?

glotz
October 18th, 2006, 10:25 AM
And what about legacy card users?

evilghost
October 18th, 2006, 08:26 PM
And what about legacy card users?

Legacy card users are only affected if RenderAccel is explicitly enabled.

Basically, this bug can be mitigated by turning off RenderAccel.

1.0.9XXX - The bug is corrected.
1.0.8XXX - RenderAccel is on by default.
1.0.7XXX - RenderAccess is off by default.

See:
http://www.nvnews.net/vbulletin/showpost.php?p=1028769&postcount=13
http://www.nvnews.net/vbulletin/showpost.php?p=1030689&postcount=15

PriceChild
October 18th, 2006, 08:44 PM
I'm expecting the fixed packages to make it to the dapper repos because it is such a huge vulnerability...

Hope so anyway ;)

evilghost
October 18th, 2006, 09:02 PM
I'm expecting the fixed packages to make it to the dapper repos because it is such a huge vulnerability...

Hope so anyway ;)

AFAIK NVIDIA doesn't have a patch for the legacy or 1.0.8XXX drivers as 1.0.9625+ patches this. When you say "fixed package" do you mean the Beta 1.0.9626? Either way, the new 1.0.9XXX drivers aren't without bugs:

http://www.nvnews.net/vbulletin/forumdisplay.php?f=14

glotz
October 18th, 2006, 09:28 PM
Legacy card users are only affected if RenderAccel is explicitly enabled.Thank you for your answer!

Helping out newbies and not pirating software? Maybe you're not so evil after all? :mrgreen:

evilghost
October 18th, 2006, 09:40 PM
Thank you for your answer!

Helping out newbies and not pirating software? Maybe you're not so evil after all? :mrgreen:

Hahaha, I'm just glad to help. :mrgreen:

jamespi
October 19th, 2006, 09:01 AM
i dont play games on ubuntu or do any 3d stuff, so i assume turning off render accel will not impact anything i do, or am i wrong? still its a bit of a drag for those that do, so hopefully nvidia will get there asses in gear and get those beta drivers finished. certainly when edgy comes out and people want to have the 3d desktop....

dr_d
October 19th, 2006, 03:03 PM
woah...

the first really big vulnerability i've ever experienced with ubuntu and it just happened to occur in the *one* piece of closed-source software in my entire system. the mind boggles.

manatawady
October 20th, 2006, 06:09 AM
Heres nVidias answer to the the questions asked about the exploit:
http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php?p_faqid=1971

Within that they say:
"We encourage users of NVIDIA graphics driver version 1.0-8762 or 1.0-8774 to upgrade to 1.0-8776, available here: http://www.nvidia.com/object/unix.html "

So rather update repo with stable release of the driver with bugfix than the beta.

Cheers, mana

ice60
October 21st, 2006, 03:37 AM
there's a patched update now. just follow the link for your system.
http://www.nvidia.com/object/unix.html

towsonu2003
October 21st, 2006, 06:42 AM
and it seems Edgy will come with an updated nvidia driver that is not vulnerable. if that happens, the fix may be backported to dapper and other supported releases.

John.Michael.Kane
October 22nd, 2006, 08:01 PM
So has anyone seen the patch ice60 linked to in the dapper/edgy repo's?

towsonu2003
October 22nd, 2006, 08:55 PM
So has anyone seen the patch ice60 linked to in the dapper/edgy repo's?

see https://launchpad.net/distros/ubuntu/+source/linux-restricted-modules-2.6.17/+bug/46034/comments/37

John.Michael.Kane
October 22nd, 2006, 09:03 PM
Thanks towsonu2003..

ebutton
October 25th, 2006, 11:35 PM
Would Ubuntu be vulnerable to this one? I don't know for certain, but I didn't risk it. When I checked my /etc/X11/xorg.conf, I see that I'm using driver "nv". I hope that means I'm safe.

Security Threat Watch
Number 154
Monday, October 23, 2006
Created for you by Network Computing & Neohapsis


This week, an interesting vulnerability in nVidia's Linux graphics
driver was released. The vulnerability could allow a remote attacker to
gain root access to any system using this driver and a carefully crafted
Web page. It is interesting because the only thing required for
exploitation is the ability to force the victim's graphics driver to
render a certain string of graphics that can be triggered by a malicious
font, flash movie or Java applet hosted by a Web page, e-mail or local
file. The most obvious avenue of remote attack is a malicious Web site
hosting the string of exploit graphics. By using a cross-site scripting
vulnerability, an attacker could even exploit victims through trusted
Web sites. Security professionals often hold the class of
vulnerabilities known as cross-site scripting, or XSS, in contempt. This
week, however, these vulnerabilities could prove to be one of the
primary exploitation methods against the nVidia driver. It is
recommended that nVidia users either install the recently released
nVidia beta drivers (version 1.0.9625) or use the default nv driver
included with the Linux kernel.

~LoKe
October 25th, 2006, 11:47 PM
Hasn't this been known since 2004? Anyways, I use the beta driver, so...

ebutton
October 26th, 2006, 12:22 AM
Hi! No, this one is new. Regretfully, I didn't check the "Servers and Security" forum until AFTER I posted. They are all over it.

I hope to have some redemption from having made reference to the xorg.conf file as a way to check one's system for the safe "nv" driver.

Good for you on being OK, too.

Regards,
EdB

John.Michael.Kane
October 27th, 2006, 12:52 AM
Heres where they stand on this issue https://launchpad.net/distros/ubuntu/+source/linux-restricted-modules-2.6.17/+bug/46034

RJARRRPCGP
November 26th, 2006, 11:10 PM
Can I be attacked just by being connected to the internet?

darkhatter
November 26th, 2006, 11:27 PM
Ubuntu is going to come with Nvidia drivers default, then news of possible exploits comes up. Compaines start producing Mac OS X virus programs news about OS X virus start popping up everywhere. Something smells fishy ;)

towsonu2003
November 27th, 2006, 02:42 AM
Seems like the fix is already released for Dapper and Edgy: https://launchpad.net/distros/ubuntu/+source/linux-restricted-modules-2.6.17/+bug/46034 (see status) so you shouldn't have any problems if you recently updated your system. Anyhow, assuming you still have the vulnerable driver:


Can I be attacked just by being connected to the internet?

probably no if you don't accept remote X connections:


It is important to note that glyph data is supplied to the X server
by the X client. Any remote X client can gain root privileges on
the X server using the proof of concept program attached.

It is also trivial to exploit this vulnerability as a DoS by causing
an existing X client program (such as Firefox) to render a long text
string. It may be possible to use Flash movies, Java applets, or
embedded web fonts to supply the custom glyph data necessary for
reliable remote code execution.

A simple HTML page containing an INPUT field with a long value is
sufficient to demonstrate the DoS.

I don't think anyone out there could send some glyph data to your X client just because you're online. But then again, I don't really know what I'm talking about...