8.1.13 dies suddenly

Moderators: Site Moderators, FAHC Science Team

Post Reply
JL_678
Posts: 37
Joined: Sun Jan 22, 2023 3:18 am

8.1.13 dies suddenly

Post by JL_678 »

Hi,

I have bene running 8.1.13 for a while now. It was a bear to setup so I have not upgraded. I have 5 nodes running, and all worked perfectly until 7/17/24, when all of them stopped getting jobs. Here is a log snippet:

Code: Select all

01:37:35:I1:OUT2483:> GET https://api.foldingathome.org/gpus HTTP/1.1
01:37:35:I3:Connecting to api.foldingathome.org:443
01:37:36:I1:OUT2483:< api.foldingathome.org:443 HTTP/1.1 200 HTTP_OK
01:37:36:E :Exception: Failed to open dynamic library 'libcuda.so': libcuda.so: cannot open shared object file: No such file or directory
01:37:36:E :Exception: Failed to open dynamic library 'libOpenCL.so': libOpenCL.so: cannot open shared object file: No such file or directory
01:37:37:I1::WU969:Requesting WU assignment
01:37:37:I1:OUT2484:> POST https://assign6.foldingathome.org/api/assign HTTP/1.1
01:37:37:I3:Connecting to assign6.foldingathome.org:443
01:37:38:I1:OUT2484:< assign6.foldingathome.org:443 HTTP/1.1 503 HTTP_SERVICE_UNAVAILABLE
01:37:38:E ::WU969:HTTP_SERVICE_UNAVAILABLE: {"error":"No appropriate assignment"}
01:37:38:I1::WU969:Retry #408 in 8 mins 32 secs
01:46:10:I1::WU969:Requesting WU assignment
01:46:10:I1:OUT2485:> POST https://assign1.foldingathome.org/api/assign HTTP/1.1
01:46:10:I3:Connecting to assign1.foldingathome.org:443
01:46:10:I1:OUT2485:< assign1.foldingathome.org:443 HTTP/1.1 503 HTTP_SERVICE_UNAVAILABLE
The cude stuff is curious because I am not running with a GPU. Not sure why it is even there. All systems are running Intel CPUs if that matters. Does this perhaps related to 8.1.3 being deprecated? I can upgrade, but it took so much work to setup 8.1.3 that I am worried about making the change.
Joe_H
Site Admin
Posts: 7922
Joined: Tue Apr 21, 2009 4:41 pm
Hardware configuration: Mac Pro 2.8 quad 12 GB smp4
MacBook Pro 2.9 i7 8 GB smp2
Location: W. MA

Re: 8.1.13 dies suddenly

Post by Joe_H »

If these systems running CPU jobs have less than 4 CPU threads available, see this topic - viewtopic.php?t=41716. The server with projects that assign WUs to CPU thread counts less than 4 is currently down.
Image

iMac 2.8 i7 12 GB smp8, Mac Pro 2.8 quad 12 GB smp6
MacBook Pro 2.9 i7 8 GB smp3
JL_678
Posts: 37
Joined: Sun Jan 22, 2023 3:18 am

Re: 8.1.13 dies suddenly

Post by JL_678 »

That explains it. Thank you.
Post Reply