Nvidia Web Driver updates for macOS High Sierra (UPDATE Nov 13, 2020)

[MICKHAEL]

18 hours ago, fullerfun said:

Anyone have any ideas?

 

all good here
10.13.6 (17G14042)
387.10.10.10.40.140

[Guest 5T33Z0]

I think, the reason is this:

 

In other words: stay offline if you want your Webdrivers to continue working!

[MICKHAEL]

okay okay. later, obviou I also got same issues))
fixed instaling https://www.nvidia.com/download/driverResults.aspx/125379/en-us
oldie but working))

who needs CUDA, also install first HighSierra version
https://www.nvidia.com/content/DriverDownload-March2009/confirmation.php?url=/Mac/Quadro_Certified/9.0.214/cudadriver-9.0.214-macos.dmg&lang=us&type=Other

Spoiler

 

 

Edited June 5, 2022 by MICKHAEL

[gorg]

Setting the clock back, staying offline, and reinstalling the 387.10.10.10.140 Web Drivers hasn't worked for me (latest High Sierra 10.13.6 build 17G14042, OpenCore 0.8.0, Nvidia GTX1080Ti, Asus X299 ROG Rampage VI Extreme). I am now on a "OConsoleUsers: gIOScreenLockState 3" loop and I can only boot with nv_disable=1.

Edited June 5, 2022 by gorg

[chris1111]

I remove certificate of this Package from Pkgutil command  but I can not test because I have no Nv card then No High Sierra 10.13

The command will downloads WebDriver-387.10.10.10.40.140.pkg then Unpack, repack

 

make sur you dont have the package in downloads folder, and Files folder, it will be deleted.
there is the script code ⬇︎

#!/bin/bash
# By chris1111

echo "Downloads WebDriver-387.10.10.10.40.140  "
echo "= = = = = = = = = = = = = = = = = = = = = = = = =  "

curl -L https://images.nvidia.com/mac/pkg/387/WebDriver-387.10.10.10.40.140.pkg -o $HOME/Downloads/WebDriver-387.10.10.10.40.140.pkg

# Expand the Packages with pkgutil
echo "Expand the Packages with pkgutil  "
echo "= = = = = = = = = = = = = = = = = = = = = = = = =  "
pkgutil --expand $HOME/Downloads/WebDriver-387.10.10.10.40.140.pkg $HOME/Downloads/Files

# Flatten the Packages with pkgutil
echo "Flatten the Packages with pkgutil  "
echo "= = = = = = = = = = = = = = = = = = = = = = = = =  "
pkgutil --flatten $HOME/Downloads/Files $HOME/Downloads/MyWebDriver-387.10.10.10.40.140.pkg

rm -rf $HOME/Downloads/WebDriver-387.10.10.10.40.140.pkg
rm -rf $HOME/Downloads/Files
echo "Done!   "
echo "= = = = = = = = = = = = = = = = = = = = = = = = =  "

attaching command ⬇︎

PKGUTIL.command.zip

 

 

 

 

 

Edited June 5, 2022 by chris1111
The command will downloads WebDriver-387.10.10.10.40.140.pkg then Unpack, repack

[gorg]

Thanks @chris1111 but that didn't help me either. After booting with nv-disable=1, disconnecting from internet and staying offline, and setting the clock back in time, I also tried to remove code signatures from everything included in the expanded Nvidia installer (kexts, bundles, preference panes) using "codesign --remove-signature" , manually copied the kexts to /S/L/E and /L/E, cleaned CRL caches in /var/db/crls, fixed permissions with Kext Utility, removed  nv-disable=1 from OpenCore, and rebooted.

 

Still no go, during boot I get "com_driver_GxxxDriver::probe falls", then the "OConsoleUsers: gIOScreenLockState 3..." loop.

Edited June 5, 2022 by gorg

[Soul92]

1 hour ago, gorg said:

Grazie @ chris1111 ma neanche questo mi ha aiutato. Dopo aver avviato con nv-disable=1, disconnesso da Internet e rimanere offline e aver riportato l'orologio indietro nel tempo, ho anche provato a rimuovere le firme del codice da tutto ciò che era incluso nel programma di installazione di Nvidia espanso (kexts, bundle, riquadri delle preferenze) usando "codesign --remove-signature" , ha copiato manualmente i kext in /S/L/E e /L/E, ha pulito le cache CRL in /var/db/crls, ha corretto i permessi con Kext Utility, ha rimosso nv-disable=1 da OpenCore, e riavviato.

 

Ancora no, durante l'avvio ottengo "com_driver_GxxxDriver::probe falls", quindi il ciclo "OConsoleUsers: gIOScreenLockState 3..." .

i also have the same problem as you, i can't solve

[fullerfun]

UPdate****
Instal new driver instead. problem solved
https://drive.google.com/uc?export=download&id=1lPuKuQqI7OtNzMMmSrQ8MqVMC1EnlfCv

 

 

Unblocked trustd and oscpd rules in little snitch.

I was then able to get it back in the stuck state by clearing the DYLD cache using ONYX. Then rebooting.

I then reboot with nv_disable=1 flag.
This allows me to boot in VESA mode. (terrible video flickering/resolution..which is what it's supposed to do)

I ran the new installer from NVidia 

https://drive.google.com/uc?export=download&id=1lPuKuQqI7OtNzMMmSrQ8MqVMC1EnlfCv
Didn't click restart, left window open.
Ran Cuda installer.

Restarted computer.

Boom, back to normal!

 

------

Here’s the best current workaround, and brought my system back to life!. (Based on solutions by eierftucht and dayo on macrumors site)

 

 

(This may not work unless you were previously running latest 10.13.6 with the latest security updates already applied. If you go to apple menu top left, select about this mac, then click system report then click on software. what version of 10.13.6 does it show? If (17G14042) then you were. If running older version, 1st run security updates, force instal of web drivers using pacifist 3.2.17, rebuild kext caches, then follow procedures  below)

 

(Also may not work unless sip is disabled)

 

 

Step 1. Physically disconnect computer from the internet. 

Step 2. Boot into Safe Mode. Everything will be extremely laggy, be patient.  (-x in clover or opencore will boot safe mode, safe mode is essential step..will not work otherwise)

Step 3. Launch Terminal and enter the command ‘sudo nano /etc/hosts’, once prompted provide the password. (You can also use something like little snitch to block trustd instead of changing hosts...in step 3 &4)

Step 4. Append the following lines to the file’s contents:

127.0.0.1 ocsp.apple.com
127.0.0.1 ocsp2.apple.com

Save changes and exit.

Step 5. Run the following batch of Terminal commands:

crlrefresh rp
sudo rm -f /var/db/crls/*cache?.db

Step 6.

Go to finder,

Make invisible files visible command-shift-.

navigate to var/db/crls/

if there is a file called "crlcache2.db" still there, throw it in the trash, enter pass if needed

navigate to /Library/Keychains/crls/ throw "valid.sqlite3" in the trash, enter pass if needed

empty the trash

 

Restart computer normally,

 

you should be back to normal, and you can now turn the internet back on without issues.

good to go!

—-

If above doesn’t work for you, a new alternative on the macrumors site, several people(credit to DTRX &  Dayo) have developed a script that will disable trustd’s functionality on your computer. Personally, I would take some caution into doing this, for security reasons, however it does work more consistently, and is a simple and fast solution.

https://forums.macrumors.com/posts/31170583

--

Alternative way is:

Restore from time machine from date before June 1st.

Completely disconnect from internet.

Boot restored system.

Open terminal. (Or use little snitch method)

Paste this, hit enter, type in pass.

sudo sh -c 'echo "0.0.0.0 ocsp.apple.com" >> /etc/hosts' && sudo sh -c 'echo "0.0.0.0 ocsp2.apple.com" >> /etc/hosts' && sudo killall -HUP mDNSResponder

Restart, reconnect to internet.

Good to go.
 

 

Edited June 15, 2022 by fullerfun

[Andres ZeroCross]

On 6/6/2022 at 8:22 AM, fullerfun said:

Here’s the best current workaround, and brought my system back to life!. (by eierftucht on macrumors site)

 

 

 

Step 1. Physically disconnect the affected device from the web. Powering down the router for a few minutes will do just fine.

Step 2. Boot into Safe Mode. Everything will be extremely laggy, be patient.  (-x in clover will boot safe mode, safe mode is essential step..will not work otherwise)

Step 3. Launch Terminal and enter the command ‘sudo nano /etc/hosts’, once prompted provide the password. (You can also use something like little snitch to block trustd instead of changing hosts...in step 3 &4)

Step 4. Append the following lines to the file’s contents:

127.0.0.1 ocsp.apple.com
127.0.0.1 ocsp2.apple.com
127.0.0.1 ocsp.digicert.com

Save changes and exit.

Step 5. Run the following batch of Terminal commands:

crlrefresh rp
sudo rm -f /var/db/crls/*cache?.db
sudo date -u 020200002020
sudo reboot

Your computer will immediately reboot after the last command. Upon seeing the desktop again, you should notice that everything is back to normal. You can now reconnect to the internet. System time and date will automatically adjust themselves upon reconnecting. If some apps throw errors related to bad time and date, another reboot will fix that. Don’t worry if you run into any scary messages upon the first reboot.

The ‘sudo date’ shift trick is 90% likely unnecessary but better safe than sorry. It’s there just to lure the system (now reverted to a clean state) into repeating any sneaky moves it’s compelled to make since the 1st of June, just to check it no longer breaks itself.

 

good to go!

 

--

Alternative way is:

Restore from time machine from date before June 1st.

Completely disconnect from internet.

Boot restored system.

Open terminal.

Paste this, hit enter, type in pass.

sudo sh -c 'echo "0.0.0.0 ocsp.apple.com" >> /etc/hosts' && sudo sh -c 'echo "0.0.0.0 ocsp2.apple.com" >> /etc/hosts' && sudo killall -HUP mDNSResponder

Restart, reconnect to internet.

Good to go.
 

 

Anyone tried this?

[Andres ZeroCross]

On 6/6/2022 at 8:22 AM, fullerfun said:

Here’s the best current workaround, and brought my system back to life!. (by eierftucht on macrumors site)

 

 

 

Step 1. Physically disconnect the affected device from the web. Powering down the router for a few minutes will do just fine.

Step 2. Boot into Safe Mode. Everything will be extremely laggy, be patient.  (-x in clover will boot safe mode, safe mode is essential step..will not work otherwise)

Step 3. Launch Terminal and enter the command ‘sudo nano /etc/hosts’, once prompted provide the password. (You can also use something like little snitch to block trustd instead of changing hosts...in step 3 &4)

Step 4. Append the following lines to the file’s contents:

127.0.0.1 ocsp.apple.com
127.0.0.1 ocsp2.apple.com
127.0.0.1 ocsp.digicert.com

Save changes and exit.

Step 5. Run the following batch of Terminal commands:

crlrefresh rp
sudo rm -f /var/db/crls/*cache?.db
sudo date -u 020200002020
sudo reboot

Your computer will immediately reboot after the last command. Upon seeing the desktop again, you should notice that everything is back to normal. You can now reconnect to the internet. System time and date will automatically adjust themselves upon reconnecting. If some apps throw errors related to bad time and date, another reboot will fix that. Don’t worry if you run into any scary messages upon the first reboot.

The ‘sudo date’ shift trick is 90% likely unnecessary but better safe than sorry. It’s there just to lure the system (now reverted to a clean state) into repeating any sneaky moves it’s compelled to make since the 1st of June, just to check it no longer breaks itself.

 

good to go!

 

--

Alternative way is:

Restore from time machine from date before June 1st.

Completely disconnect from internet.

Boot restored system.

Open terminal.

Paste this, hit enter, type in pass.

sudo sh -c 'echo "0.0.0.0 ocsp.apple.com" >> /etc/hosts' && sudo sh -c 'echo "0.0.0.0 ocsp2.apple.com" >> /etc/hosts' && sudo killall -HUP mDNSResponder

Restart, reconnect to internet.

Good to go.
 

 

I can confirm, this trick is working good

[RaymondK]

On 6/5/2022 at 9:22 PM, fullerfun said:

Here’s the best current workaround, and brought my system back to life!. (by eierftucht on macrumors site)

 

 

 

Step 1. Physically disconnect the affected device from the web. Powering down the router for a few minutes will do just fine.

Step 2. Boot into Safe Mode. Everything will be extremely laggy, be patient.  (-x in clover will boot safe mode, safe mode is essential step..will not work otherwise)

Step 3. Launch Terminal and enter the command ‘sudo nano /etc/hosts’, once prompted provide the password. (You can also use something like little snitch to block trustd instead of changing hosts...in step 3 &4)

Step 4. Append the following lines to the file’s contents:

127.0.0.1 ocsp.apple.com
127.0.0.1 ocsp2.apple.com
127.0.0.1 ocsp.digicert.com

Save changes and exit.

Step 5. Run the following batch of Terminal commands:

crlrefresh rp
sudo rm -f /var/db/crls/*cache?.db
sudo date -u 020200002020
sudo reboot

Your computer will immediately reboot after the last command. Upon seeing the desktop again, you should notice that everything is back to normal. You can now reconnect to the internet. System time and date will automatically adjust themselves upon reconnecting. If some apps throw errors related to bad time and date, another reboot will fix that. Don’t worry if you run into any scary messages upon the first reboot.

The ‘sudo date’ shift trick is 90% likely unnecessary but better safe than sorry. It’s there just to lure the system (now reverted to a clean state) into repeating any sneaky moves it’s compelled to make since the 1st of June, just to check it no longer breaks itself.

 

good to go!

 

--

Alternative way is:

Restore from time machine from date before June 1st.

Completely disconnect from internet.

Boot restored system.

Open terminal.

Paste this, hit enter, type in pass.

sudo sh -c 'echo "0.0.0.0 ocsp.apple.com" >> /etc/hosts' && sudo sh -c 'echo "0.0.0.0 ocsp2.apple.com" >> /etc/hosts' && sudo killall -HUP mDNSResponder

Restart, reconnect to internet.

Good to go.
 

 

Thank as it works great again.

[gorg]

Why WebDriver-387.10.10.15.15.108 is nowhere to be found in all Internet ?

[gorg]

43 minutes ago, Hervé said:

You can't have searched much... The link on p1 is just corrupt. Use this:

https://images.nvidia.com/mac/pkg/387/WebDriver-387.10.10.10.40.108.pkg

I have and I do have 387.10.10.10.40.108, alongside older ones archived for safe keeping. But not 387.10.10.15.15.108, which suspiciouly was supposed to be the only (?) one to be HS build number independent (?).

Edited June 11, 2022 by gorg

[startergo]

The cat is out of the bag

https://drive.google.com/open?id=1lPuKuQqI7OtNzMMmSrQ8MqVMC1EnlfCv

[user01100]

On 6/11/2022 at 4:57 AM, gorg said:

Why WebDriver-387.10.10.15.15.108 is nowhere to be found in all Internet ?

https://egpu.io/forums/thunderbolt-enclosures/nvidia-titan-v-gpu-for-deep-learning-ai-osx-10-13-3-10-13-4/paged/3/#post-46382

[Guest 5T33Z0]

@fullerfun As far as editing the etc/hosts file is concerned: there's an App for that called GasMask, so you don't need to do it in safe mode to do it.

 

And also: can you fix the formatitin of your post so that it's actually readable?

 

Edited June 18, 2022 by 5T33Z0

[supergullit]

Hello

I formatted my Mac Pro 3.1 and using Dosdude tool I Installed High Sierra, the version is 10.13.6 (17G66). There is no web driver for this version so it uses the default driver (some tasks are very slow). The card is a NVIDIA GeForce GTX 650 Ti (in addition to the default 2600 XT).

I prefer not to update high sierra to the latest version (2020-005 update) because it becomes very unstable with frequent and sudden logout, I've already tried a couple of times.

Is there no way I can install a compatible version of the web drivers?
Thanks

[fantomas]

@supergullit

 

Hi

 

Nvidia Web Driver - 387.10.10.10.40.105 --> build 17G65 should make the job but you'll have to tweak it a little bit to skip hardware/software check.

I've made in the past a "how to" video that could be helpful. 

 

 

 

BR

 

fantomas

[treyhamilton25]

Ok I have been racking my brain forever. I have the GTX 1050ti. Webdrivers installed, laptop sees the card, but when I boot the system it's booting under my integrated graphics instead of my 1050Ti. I have the bootarg set to use nvidia, but it doesn't. When I click the Apple logo and go to "About this Mac" I just see the regular HD graphics. I'm running OC and followed the guide there for install. 

[1Revenger1]

Often the internal display is only connected to the iGPU, and Optimus is used to copy over data from the dGPU to display on to the iGPU. Unless you have some firmware options which change this, your likely out of luck. Sometimes the HDMI/display outs are only connected to the dGPU though, so you may see it appear for those ports if plugged into.

[treyhamilton25]

I know that during the setup I had to set ig_platform_id set for HD630 because I needed the webdrivers post install. Was just hoping to get ndivia working. I know with my previous laptop (Hp ENVY 17) I was able to get dGPU running instead of iGPU. Everything worked flawlessly. Just figured I could do the same with this one too. (Hp Omen 17t) I'll just keep doing more research I suppose.