Trouble installing on Ubuntu & some other issues

[Asgaroth]

Hi,

The post here is for configuring the fah client from cli, see if you can get it started from cli following those steps.
The post is for arch linux, but the configuration for fah client config is the same. I'm not sure of the gui client issue on ubuntu, but you may be able to get it configured through the cli by editing the config file directly. just make sure fah client is properly shut dowb before editing the config client

[Azmodes]

Thanks, but right now the issue isn't configuration, I can do that just fine through the terminal (and have for a number of other Ubuntu systems that are now happily folding). It's that after I do, it doesn't see any GPUs to crunch with and can't connect to a required server.

Code: Select all

[..]
15:34:53:         OS: Linux 5.3.0-29-generic x86_64
15:34:53:    OS Arch: AMD64
15:34:53:       GPUs: 0
15:34:53:       CUDA: 7.5
15:34:53:CUDA Driver: 10020
15:34:53:***********************************************************************
15:34:53:<config>
15:34:53:  <!-- Slot Control -->
15:34:53:  <power v='full'/>
15:34:53:
15:34:53:  <!-- Folding Slots -->
15:34:53:  <slot id='0' type='GPU'/>
15:34:53:  <slot id='1' type='GPU'/>
15:34:53:</config>
15:34:53:Connecting to assign-GPU.stanford.edu:32563
15:37:03:WARNING:Attempting to update GPUs.txt from assign-GPU.stanford.edu:80: Failed to connect to assign-GPU.stanford.edu:32563: Connection timed out
15:37:03:Connecting to assign-GPU2.stanford.edu:32563
15:37:03:WARNING:Attempting to update GPUs.txt from assign-GPU2.stanford.edu:80: Failed to connect to assign-GPU2.stanford.edu:32563: Connection refused
[..]

[Asgaroth]

I wonder if their servers are still being overloaded with the influx of new donars, they are working on correcting the issue, but your client on that system should keep trying.
As for it not detecting your gpu's, i'm assuming you've tried to stop the client and restart it. I can see that it's detected no gpu's there but your config has 2 slots assigned for gpu's.

I'm not too familiar with the ubuntu/debian variant(s), but did you install the opencl libraries?

What does

Code: Select all

FAHClient --info

give you?

[Azmodes]

I wonder if their servers are still being overloaded with the influx of new donars, they are working on correcting the issue, but your client on that system should keep trying.
As for it not detecting your gpu's, i'm assuming you've tried to stop the client and restart it. I can see that it's detected no gpu's there but your config has 2 slots assigned for gpu's.

Well, all my other hosts where the cards are being detected are getting a (reasonably enough) steady stream of work. Yes, tried restarting. Those two slots I added manually to the config file, it didn't do that on its own.

I'm not too familiar with the ubuntu/debian variant(s), but did you install the opencl libraries?

Yes. All those computers are normally running various BOINC projects, most of which require OCL. No issues with those.

What does

Code: Select all

FAHClient --info

give you?

Assorted information about the system, including 0 GPUs.

[Asgaroth]

Does the "assorted information" show that it detects 2 cuda gpu's and 2 opencl gpu's?

Something like this:

Code: Select all

   CUDA Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:6.1 Driver:10.0
   CUDA Device 1: Platform:0 Device:1 Bus:11 Slot:0 Compute:6.1 Driver:10.0
   CUDA Device 2: Platform:0 Device:2 Bus:13 Slot:0 Compute:6.1 Driver:10.0
   CUDA Device 3: Platform:0 Device:3 Bus:14 Slot:0 Compute:6.1 Driver:10.0
   CUDA Device 4: Platform:0 Device:4 Bus:15 Slot:0 Compute:6.1 Driver:10.0
   CUDA Device 5: Platform:0 Device:5 Bus:17 Slot:0 Compute:6.1 Driver:10.0
OpenCL Device 0: Platform:1 Device:0 Bus:1 Slot:0 Compute:1.2 Driver:415.22
OpenCL Device 1: Platform:1 Device:1 Bus:11 Slot:0 Compute:1.2 Driver:415.22
OpenCL Device 2: Platform:1 Device:2 Bus:13 Slot:0 Compute:1.2 Driver:415.22
OpenCL Device 3: Platform:1 Device:3 Bus:14 Slot:0 Compute:1.2 Driver:415.22
OpenCL Device 4: Platform:1 Device:4 Bus:15 Slot:0 Compute:1.2 Driver:415.22
OpenCL Device 5: Platform:1 Device:5 Bus:17 Slot:0 Compute:1.2 Driver:415.22

[Azmodes]

Nope, the only GPU-related entries are

Code: Select all

GPUs: 0
CUDA: 6.1
CUDA driver: 10010

[Asgaroth]

Hmm, I wonder if the client is unable to use the installed cuda/ocl libraries, but, you say the client is working on some of your other systems, are these other systems running the same version of ubuntu as the one you're having issues with?

[Azmodes]

The ones working are all Ubuntu 18.04, the ones not 19.xx. I was also thinking that that's somehow related, but I'm not IT savvy enough to offer any theories.

[rixon]

So which CUDA do you have installed, 6.1 or 7.5? Are those installed on the other systems (since I don't know which CUDA versions are supported.) Are you using something like Environment Modules to switch your CUDAs, and if so, do you have a newer release installed you can run from?

[None4U]

Try deleting GPUs.txt and restarting.

[Azmodes]

So which CUDA do you have installed, 6.1 or 7.5? Are those installed on the other systems (since I don't know which CUDA versions are supported.) Are you using something like Environment Modules to switch your CUDAs, and if so, do you have a newer release installed you can run from?

Ah yeah, sorry. Those are actually two different systems, but the issue is the same on both. One has two 2070S, the other a 1060 6GB + 1030, hence the difference in CUDA versions.

Try deleting GPUs.txt and restarting.

I can give that a try later today.

[Azmodes]

Try deleting GPUs.txt and restarting.

I can give that a try later today.

No results found.