Message boards : Puzzles and science : FAROS
Message board moderation
| Author | Message |
|---|---|
rilianSend message Joined: 28 Sep 26 Posts: 261 Credit: 43,482,494 RAC: 474,864 |
|
rilianSend message Joined: 28 Sep 26 Posts: 261 Credit: 43,482,494 RAC: 474,864 |
|
|
25 years old Send message Joined: 27 Sep 26 Posts: 349 Credit: 5,210,013 RAC: 137,017 |
Yes. These are actual work units (WUs), but they are not reported to anyone; they are used solely for debugging purposes—meaning they are independent of any projects.
|
Tamagoch47 years old Send message Joined: 29 Sep 26 Posts: 68 Credit: 14,373,389 RAC: 423,227 |
|
|
51 years old Send message Joined: 28 Sep 26 Posts: 67 Credit: 45,851,701 RAC: 928,453 |
I just tried to download a bunch of Faros tasks for Nvidia, Linux but the download failed of all of them. ChatGPT had this to say: Faros v2 downloads are failing because BOINC rejects the application binary: Application file faros_v2_linux_x86_64 missing signature BOINC cannot accept this file Checksum or signature error for faros_v2_linux_x86_64 followed by: check_file_signature: decrypt_public error The Faros input/SFT files download normally, so this appears to be a server-side app signing problem with faros_v2_linux_x86_64. |
Steve DoddSend message Joined: 28 Sep 26 Posts: 27 Credit: 53,008,805 RAC: 218,076 |
Ditto |
ConanSend message Joined: 29 Sep 26 Posts: 38 Credit: 19,720,099 RAC: 239,916 |
I have yet to see a work unit let alone a failed one. Conan |
|
25 years old Send message Joined: 27 Sep 26 Posts: 349 Credit: 5,210,013 RAC: 137,017 |
I have made changes to the application; now we will see what the errors are.
|
Tamagoch47 years old Send message Joined: 29 Sep 26 Posts: 68 Credit: 14,373,389 RAC: 423,227 |
|
|
25 years old Send message Joined: 27 Sep 26 Posts: 349 Credit: 5,210,013 RAC: 137,017 |
we need more tester
|
|
51 years old Send message Joined: 28 Sep 26 Posts: 67 Credit: 45,851,701 RAC: 928,453 |
I had several of my hosts at Faros`disposal yesterday and today but they didn`t get any work. Although left at idle. Which hardware are you targeting now? Numberfields and Einstein run CUDA, right? Also AMD or cpu? Is there a limit of tasks per host, when they have proven to work successfuly? |
|
25 years old Send message Joined: 27 Sep 26 Posts: 349 Credit: 5,210,013 RAC: 137,017 |
There are no limits to the scope of the work. It involves all types of devices. I compare driver crashes and different kernels.
|
|
51 years old Send message Joined: 28 Sep 26 Posts: 67 Credit: 45,851,701 RAC: 928,453 |
So why were my hosts not getting any although the server page showed about 50 tasks ready to send. Some of the tasks my GPUs worked on yesterday were definitely GPU-only. You didnd`t plan to send those out to a cpu, did you? |
|
Send message Joined: 28 Sep 26 Posts: 14 Credit: 20,207,102 RAC: 1,209,823 |
I tried grabbing CPU and GPU work and also received none. |
|
25 years old Send message Joined: 27 Sep 26 Posts: 349 Credit: 5,210,013 RAC: 137,017 |
In Faros, CPU and GPU tasks run on separate architectures. The disk is currently full; I will add a new disk to the server within an hour. This will require a brief maintenance period, after which everything will return to normal. Users are currently experiencing issues with file uploads and downloads. It appears that some users have downloaded a large number of work units (WUs) but abandoned the project without completing any of them; I will prepare a Python script to identify these instances and return the affected tasks to the work queue.
|
|
51 years old Send message Joined: 28 Sep 26 Posts: 67 Credit: 45,851,701 RAC: 928,453 |
Bug report: Faros v0.06 OpenCL task fails after BOINC suspend/resume on Windows A Faros GPU task ran normally but failed immediately when restarted after pausing and resuming BOINC. The same computer continues to process other Faros v0.06 tasks, suggesting a problem related to suspend/resume. Environment: - Application: Faros v0.06 (opencl_ati), Windows x86_64 - Executable: faros_v6_windows_x86_64.exe - BOINC: 8.2.4 - GPU reported by application: AMD gfx1150 - OpenCL platform: AMD Accelerated Parallel Processing - Host ID: 156 Affected task: - Result ID: 7990087 - Workunit ID: 7739767 - Task name: faros_cd26653c45af_1792570867_1848b70_0 - Exit status: 1 - Outcome: Computation error Observed behavior: The initial run started on October 22 at 00:30:38 and successfully selected the GPU/OpenCL backend: Search FstatMethod used: 'ResampGPU' GPU Backend used for Search/Recalc and/or semi coherent step: 'OpenCL' Input validation succeeded, and computation progressed to counter 1924, with periodic "c" markers in the log. After BOINC was paused and resumed, the application restarted on October 23 at 09:40:39 with a new process ID. It then failed immediately: XLAL Error - MAIN (HierarchSearchGCT.c:1009): Generic Backend not implemented yet. XLAL Error - MAIN (HierarchSearchGCT.c:1009): Function not implemented The application subsequently reported: ERROR: MAIN() returned with error '1' called boinc_finish(1) Expected behavior: After resuming BOINC, the task should select the OpenCL backend again and continue from its saved checkpoint. Possible cause: The restart appears to enter an unsupported Generic backend path instead of the previously working OpenCL path. The failure occurs before any checkpoint-loading message. The exact reason for this backend selection is unknown. This has been observed once; reproducibility has not yet been established. Workunit: https://bitboinc.athena.org.tr/workunit.php?wuid=7739767 Result: https://bitboinc.athena.org.tr/result.php?resultid=7990087 |
[SG]KidDosCrunchSend message Joined: 8 Oct 26 Posts: 29 Credit: 87,709,633 RAC: 1,721,339 |
Meanwhile I have 100 Faros WU succesfully returned. It is still showing the Flathead Badge? Why? |
[SG]KidDosCrunchSend message Joined: 8 Oct 26 Posts: 29 Credit: 87,709,633 RAC: 1,721,339 |
In reply to [SG]KidDosCrunch's message of 24 Oct 2026: Meanwhile I have 100 Faros WU succesfully returned. It is still showing the Flathead Badge? Why? Now it is OK. |
ConanSend message Joined: 29 Sep 26 Posts: 38 Credit: 19,720,099 RAC: 239,916 |
Still haven't received a single WU, but I suppose I will get some eventually, I wait. Conan |
ConanSend message Joined: 29 Sep 26 Posts: 38 Credit: 19,720,099 RAC: 239,916 |
Have been able to finally report my first ever FAROS work units only to have them all fail with validate errors. Was unable to report for two days due to server migration, so not sure if that was the issue. So I still have no successful work units to show against my name. I wait some more. Conan |