This is in the Event Log when starting BOINC:

CUDA: NVIDIA GPU 0: GeForce GTX 1060 6GB (driver version 390.11, CUDA version 9.1, compute capability 6.1, 4096MB, 3988MB available, 4503 GFLOPS peak)
OpenCL: NVIDIA GPU 0: GeForce GTX 1060 6GB (driver version 390.116, device version OpenCL 1.2 CUDA, 6078MB, 3988MB available, 4503 GFLOPS peak)

So is this:

SETI@homeFound app_config.xml
SETI@homeEntry in app_config.xml for app 'astropulse_v7', plan class 'opencl_nvidia_100' doesn't match any app versions

I have no clue. That plan class matches what I use for my astropulse app in my app_config. Is the astropulse app marked executable? I have

astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100 2.5 MB (2,508,192 bytes) Fri 18 Sep 2015 01∶50∶36 AM PDT

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

That plan class matches what I use for my astropulse app in my app_config. Is the astropulse app marked executable? I have

astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100 2.5 MB (2,508,192 bytes) Fri 18 Sep 2015 01∶50∶36 AM PDT

And check that the name is exactly the same as what is in the app_info.xml file.

---

Grant
Darwin NT

I have the same executable:

-rwxr-xr-x 1 boinc boinc  2508192 May  2 05:08 astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100

I tried editing app_config.xml and re-entering info JIC there was an invisible character, no change.

---

I don't have an app name in app_config.xml. I thought the app_class would default to the proper app. Can I override app_class by using the app name instead? I assume that an app update would break app_config, no big deal to re-edit it.

---

<app_config>
 <app>
  <name>setiathome_v8</name>
  <gpu_versions>
  <gpu_usage>1.0</gpu_usage>
  <cpu_usage>1.0</cpu_usage>
  </gpu_versions>
 </app>
 <app>
  <name>astropulse_v7</name>
  <gpu_versions>
  <gpu_usage>0.5</gpu_usage>
  <cpu_usage>1.0</cpu_usage>
  </gpu_versions>
 </app>
</app_config>

Works on my systems.

---

Grant
Darwin NT

Well the last AP in my list has just gone.

Cheers.

Well the last AP in my list has just gone.

I've got one that timeout on one wingman, and now waiting for the timeout from the following one.
Should be gone by the end of this month with luck- 2 months after it first went out.

---

Grant
Darwin NT

That plan class matches what I use for my astropulse app in my app_config. Is the astropulse app marked executable? I have

astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100 2.5 MB (2,508,192 bytes) Fri 18 Sep 2015 01∶50∶36 AM PDT

And check that the name is exactly the same as what is in the app_info.xml file.

Yes, check for double length underscores as those are easy to miss.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

2 more resends for the 1st of June.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I'll soon be down to 0.
A host that appeared to have gone AWOL made a surprising return & my last remaining AP task has validated. Once the database assimilates, deletes & purges it, I won't have any AP WUs left on either of my systems Tasks lists.

---

Grant
Darwin NT

With no AP work being split out of those last Arecibo tapes and the number of them out in the field dropping I'm starting to wonder if something is being planned for Marvin.

That's just speculation of cause. ;-)

Cheers.

I still have 3 AP tasks in the database. Nothing for today.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I found the problem.

A snippet from my app_config.xml:

<app_name>astropulse_v7</app_name>
<plan_class>opencl_nvidia_100</plan_class>

A snippet from my client_state.xml:

<app_name>astropulse_v7</app_name>
<plan_class>opencl_nvidia_cc1</plan_class>
<file_name>astropulse_7.08_x86_64-pc-linux-gnu__opencl_nvidia_100</file_name>

I changed app_config.xml to use opencl_nvidia_cc1 and BOINC started up without the error. I thought the filename's ending opencl_nvidia_100 was the plan_class, was I wrong?

This begs the question: why don't others with the same app_config.xml get the error? Because this host was setup about two months ago?

---

I changed app_config.xml to use opencl_nvidia_cc1 and BOINC started up without the error. I thought the filename's ending opencl_nvidia_100 was the plan_class, was I wrong?

This begs the question: why don't others with the same app_config.xml get the error? Because this host was setup about two months ago?

Yes. That host was obviously never set up as a anonymous platform from the beginning so it had the several flavors of apps sent by the schedulers in the client_state. It just happened to have the cc1 version instead of the normal opencl_nvidia_100 version which we all have standardized on with anonymous.

Good job sleuthing it out. If you don't have any AP tasks, you can always change the client_state to the opencl_nvidia_100 plan class for conformity.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Now less than 600 AP WUs in progress.

---

Grant
Darwin NT

With no AP work being split out of those last Arecibo tapes and the number of them out in the field dropping I'm starting to wonder if something is being planned for Marvin.

That's just speculation of cause. ;-)

I'm still going with some kind of server thingamabob being planned.

Cheers.

I'm hoping they give us more APs after today's outage. If I only had a few, I'd save them for after a long outage.

No APs for us :-( . The aricebo schedule doesn't have them recording any data for us until June 18 and the way they have been modifying the schedule doesn't make me hopeful that even then they will give us any. I would love a science update on the status of data from Aricebo and what is the status of that project.

When shall I be eligible to receive Astropulse Tasks?
My last AP's were on early May with only 5 tasks, on April I received only 13...
I have 13 active Computers for this Project and this is the first time I have ZERO tasks since 1 month...
Any Explanation??? Anybody???

They haven't split any AP tasks for several weeks. Only resends trickling out. Nobody has any AP tasks, there are only 532 tasks out in the field now.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)