Hashcat Newbie Multi GPU Setup
#1
Hashcat Newbie needs help. I have an I3-9100F, with 16GB Ram, 240GB SSD, and One GTX 2060 Super currently installed. ( I have 2 more GTX 2060 Super's ready to install). My installed OS is Ubuntu 20.04 Desktop. I have applied the Linux Hashcat timeout patch as per https://hashcat.net/wiki/doku.php?id=timeout_patch

Before applying the patch:

hashcat (v6.1.1-120-g15bf8b730) starting...

* Device #1: WARNING! Kernel exec timeout is not disabled.
            This may cause "CL_OUT_OF_RESOURCES" or related errors.
            To disable the timeout, see: https://hashcat.net/q/timeoutpatch
* Device #2: WARNING! Kernel exec timeout is not disabled.
            This may cause "CL_OUT_OF_RESOURCES" or related errors.
            To disable the timeout, see: https://hashcat.net/q/timeoutpatch
CUDA API (CUDA 11.1)
====================
* Device #1: GeForce RTX 2060 SUPER, 7681/7979 MB, 34MCU

OpenCL API (OpenCL 1.2 CUDA 11.1.102) - Platform #1 [NVIDIA Corporation]
========================================================================
* Device #2: GeForce RTX 2060 SUPER, skipped

Minimum password length supported by kernel: 8
Maximum password length supported by kernel: 63

Hashes: 4 digests; 4 unique digests, 3 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates

Applicable optimizers applied:
* Zero-Byte
* Brute-Force
* Slow-Hash-SIMD-LOOP

Watchdog: Temperature abort trigger set to 90c

Host memory required for this attack: 597 MB


After patching: 

hashcat (v6.1.1-120-g15bf8b730) starting...

CUDA API (CUDA 11.1)
====================
* Device #1: GeForce RTX 2060 SUPER, 7681/7979 MB, 34MCU

OpenCL API (OpenCL 1.2 CUDA 11.1.102) - Platform #1 [NVIDIA Corporation]
========================================================================
* Device #2: GeForce RTX 2060 SUPER, skipped

Minimum password length supported by kernel: 8
Maximum password length supported by kernel: 63

Hashes: 4 digests; 4 unique digests, 3 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates

Applicable optimizers applied:
* Zero-Byte
* Brute-Force
* Slow-Hash-SIMD-LOOP

Watchdog: Temperature abort trigger set to 90c

Host memory required for this attack: 597 MB


So if I am reading this right, the patch has been applied.

Now when I put in my second GTX 2060 Super, I get: 

hashcat (v6.1.1-120-g15bf8b730) starting...

* Device #2: WARNING! Kernel exec timeout is not disabled.
            This may cause "CL_OUT_OF_RESOURCES" or related errors.
            To disable the timeout, see: https://hashcat.net/q/timeoutpatch
* Device #4: WARNING! Kernel exec timeout is not disabled.
            This may cause "CL_OUT_OF_RESOURCES" or related errors.
            To disable the timeout, see: https://hashcat.net/q/timeoutpatch
CUDA API (CUDA 11.1)
====================
* Device #1: GeForce RTX 2060 SUPER, 7653/7979 MB, 34MCU
* Device #2: GeForce RTX 2060 SUPER, 7881/7982 MB, 34MCU

OpenCL API (OpenCL 1.2 CUDA 11.1.102) - Platform #1 [NVIDIA Corporation]
========================================================================
* Device #3: GeForce RTX 2060 SUPER, skipped
* Device #4: GeForce RTX 2060 SUPER, skipped

Minimum password length supported by kernel: 8
Maximum password length supported by kernel: 63

Hashes: 4 digests; 4 unique digests, 3 unique salts
Bitmaps: 16 bits, 65536 entries, 0x0000ffff mask, 262144 bytes, 5/13 rotates

Applicable optimizers applied:
* Zero-Byte
* Brute-Force
* Slow-Hash-SIMD-LOOP

Watchdog: Temperature abort trigger set to 90c

Host memory required for this attack: 1194 MB

If I am reading this correctly. My primary 2060 card is error free, but my secondary 2060 is not. So my questions are: 


1) Does it matter if I am using Desktop Ubuntu, or do I need to use server?
2) How do I edit this patch to apply to secondary video cards? (I would eventually like to have 6-8 identical cards)
Reply
#2
Both cards have been detected:
Code:
CUDA API (CUDA 11.1)
====================
* Device #1: GeForce RTX 2060 SUPER, 7653/7979 MB, 34MCU
* Device #2: GeForce RTX 2060 SUPER, 7881/7982 MB, 34MCU

1) Does it matter if I am using Desktop Ubuntu, or do I need to use server?
Same.
2) How do I edit this patch to apply to secondary video cards? (I would eventually like to have 6-8 identical cards)
Do not edit. Cards are already detected.

Please post output of :
hashcat -I
and hashcat -b -m 0
Reply
#3
(11-01-2020, 01:16 PM)Mem5 Wrote: Both cards have been detected:

Code:
CUDA API (CUDA 11.1)

====================

* Device #1: GeForce RTX 2060 SUPER, 7653/7979 MB, 34MCU

* Device #2: GeForce RTX 2060 SUPER, 7881/7982 MB, 34MCU



1) Does it matter if I am using Desktop Ubuntu, or do I need to use server?

Same.

2) How do I edit this patch to apply to secondary video cards? (I would eventually like to have 6-8 identical cards)

Do not edit. Cards are already detected.



Please post output of :

hashcat -I

and hashcat -b -m 0




Output of hashcat -I


hashcat (v6.1.1-120-g15bf8b730) starting...

CUDA Info:
==========

CUDA.Version.: 11.1

Backend Device ID #1 (Alias: #3)
  Name...........: GeForce RTX 2060 SUPER
  Processor(s)...: 34
  Clock..........: 1665
  Memory.Total...: 7979 MB
  Memory.Free....: 7691 MB

Backend Device ID #2 (Alias: #4)
  Name...........: GeForce RTX 2060 SUPER
  Processor(s)...: 34
  Clock..........: 1665
  Memory.Total...: 7982 MB
  Memory.Free....: 7881 MB

OpenCL Info:
============

OpenCL Platform ID #1
  Vendor..: NVIDIA Corporation
  Name....: NVIDIA CUDA
  Version.: OpenCL 1.2 CUDA 11.1.102

  Backend Device ID #3 (Alias: #1)
    Type...........: GPU
    Vendor.ID......: 32
    Vendor.........: NVIDIA Corporation
    Name...........: GeForce RTX 2060 SUPER
    Version........: OpenCL 1.2 CUDA
    Processor(s)...: 34
    Clock..........: 1665
    Memory.Total...: 7979 MB (limited to 1994 MB allocatable in one block)
    Memory.Free....: 0 MB
    OpenCL.Version.: OpenCL C 1.2
    Driver.Version.: 455.32.00

  Backend Device ID #4 (Alias: #2)
    Type...........: GPU
    Vendor.ID......: 32
    Vendor.........: NVIDIA Corporation
    Name...........: GeForce RTX 2060 SUPER
    Version........: OpenCL 1.2 CUDA
    Processor(s)...: 34
    Clock..........: 1665
    Memory.Total...: 7982 MB (limited to 1995 MB allocatable in one block)
    Memory.Free....: 0 MB
    OpenCL.Version.: OpenCL C 1.2
    Driver.Version.: 455.32.00


the output of hashcat -b -m 0 is:

hashcat (v6.1.1-120-g15bf8b730) starting in benchmark mode...

Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.

* Device #2: WARNING! Kernel exec timeout is not disabled.
            This may cause "CL_OUT_OF_RESOURCES" or related errors.
            To disable the timeout, see: https://hashcat.net/q/timeoutpatch
* Device #4: WARNING! Kernel exec timeout is not disabled.
            This may cause "CL_OUT_OF_RESOURCES" or related errors.
            To disable the timeout, see: https://hashcat.net/q/timeoutpatch
CUDA API (CUDA 11.1)
====================
* Device #1: GeForce RTX 2060 SUPER, 7695/7979 MB, 34MCU
* Device #2: GeForce RTX 2060 SUPER, 7881/7982 MB, 34MCU

OpenCL API (OpenCL 1.2 CUDA 11.1.102) - Platform #1 [NVIDIA Corporation]
========================================================================
* Device #3: GeForce RTX 2060 SUPER, skipped
* Device #4: GeForce RTX 2060 SUPER, skipped

Benchmark relevant options:
===========================
* --optimized-kernel-enable

Hashmode: 0 - MD5


* At this point the cursur flashes 9 times, then just hangs
Reply
#4
Hi, Did You find any solutions to the problem?
I have this problem too.
Reply
#5
Hi Not sure if you fixed this. I had the same problem and no one helped so I tried making a copy of the “20-nvidia.conf” file in /usr/share/X11/xorg.conf.d/ and paste it back in the same location. You now have “20-nvidia(copy1).conf” as well as the original. Then restart the system.
I'm not sure if this is the correct fix but I no longer have the error. I'm guessing you need a file for each card if so.
Reply