For the last several months I have not gotten any CPU work - only GPU for my nVidia 670 card (CUDA 50 work units).

I am using the lunatics optimized exe and I believe it ran fine for several months and then sometime - just before Christmas 2013 all CPU work stopped. No messages in BIONC, updated BOINC to the newest version about a month ago (to 7.2.33), no idea what else to check. I have not touched my preferences on the site for many months before I had the issue.

What do I need to check to get CPU work?

Thanks,
Brian

---

In your event log do your request looks like this:
SETI@home Requesting new tasks for NVIDIA
or like this:
SETI@home Requesting new tasks for CPU and NVIDIA

You can check your SETI@home preferences to make sure you have CPU work enabled.

EDIT: Looking at your errors You got 49 CPU tasks on the 27th but they ended in error.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Looks like you received some yesterday but they all errored out with about 3 second run times - Exit status -226 (0xffffffffffffff1e) ERR_TOO_MANY_EXITS.

Someone can probably help with that error, I'm not sure I'm right. In the meantime the server has limited you to one S@H v7 cpu task per day. That will increase when you can send in some good results.

---

Another Fred
Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop.

Probably means he's installed an incompatible app, he should try rerunning the Lunatics Installer, and trying a different app, preferably an AMD compatible app.

Claggy

For trouble shooting purposes, set your screensaver
to 'None' if you're using one. Check that your antivirus program is not interfering in any way. Please post the first 40 lines of the event logs.
Make sure your cpu is not overheating due to heavy use.

That should do for a start.

This is the only log I can really find - job_log_setiathome.berkeley.edu.txt - in the Boinc\Data folder, Everything in it is garbage:

"1361568632 ue 18589.545837 ct 78.905310 fe 47560000000000 nm 16oc12ab.8654.6041.4.10.155_0 et 195.248341
1361568806 ue 18589.545837 ct 60.668790 fe 47560000000000 nm 16oc12ab.8654.6041.4.10.167_0 et 172.890102
"

No screensavers set and the CPU is not overheating (currently running at about 34 C). CPU work is enabled - I guess I will reinstall the Lunatics app again.

---

This is the only log I can really find - job_log_setiathome.berkeley.edu.txt - in the Boinc\Data folder, Everything in it is garbage:

"1361568632 ue 18589.545837 ct 78.905310 fe 47560000000000 nm 16oc12ab.8654.6041.4.10.155_0 et 195.248341
1361568806 ue 18589.545837 ct 60.668790 fe 47560000000000 nm 16oc12ab.8654.6041.4.10.167_0 et 172.890102
"

No screensavers set and the CPU is not overheating (currently running at about 34 C). CPU work is enabled - I guess I will reinstall the Lunatics app again.

The event log opens in a separate window. In BOINC Manager, pull down the advanced menu and select "activity log". Post the first 30-40 lines, down to where it starts tasks.

Priority should be to rerun the installer and make sure you select an app suitable to your cpu.

Looks like scheduler gave you 14 cpu tasks today, despite that daily limit of one. Have any run yet?

Edit: I see now that most have errored out, same symptoms as yesterday.

---

Another Fred
Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop.

In the Activity log - the first 40 or so lines are Starting and uploading tasks but I went father up to get the preferences:

3/2/2014 7:44:08 AM | SETI@home | Sending scheduler request: To fetch work.
3/2/2014 7:44:08 AM | SETI@home | Reporting 32 completed tasks
3/2/2014 7:44:08 AM | SETI@home | Requesting new tasks for CPU and NVIDIA
3/2/2014 7:44:11 AM | SETI@home | Scheduler request completed: got 0 new tasks
3/2/2014 7:44:11 AM | SETI@home | No tasks sent
3/2/2014 7:44:11 AM | SETI@home | No tasks are available for SETI@home Enhanced
3/2/2014 7:44:11 AM | SETI@home | No tasks are available for AstroPulse v6
3/2/2014 7:44:11 AM | SETI@home | Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them
3/2/2014 7:44:11 AM | SETI@home | Tasks for Intel GPU are available, but your preferences are set to not accept them
3/2/2014 7:44:11 AM | SETI@home | This computer has finished a daily quota of 1 tasks
3/2/2014 7:45:53 AM | SETI@home | Computation for task 24se13ad.24872.4157.438086664208.12.224_1 finished
3/2/2014 7:45:53 AM | SETI@home | Starting task 27mr13ac.15924.8656.438086664199.12.223_1 using setiathome_v7 version 700 (cuda50) in slot 0

I reran the installer last night, The only options for SSE or SSE 2 apps - nothing specific to Intel or AMD. THe Phenom I have is SSE 1 - 4a si that should not be the issue. I am choosing the 64 bit installer and I redownloaded the installer last night as well.

---

I reran the installer last night, The only options for SSE or SSE 2 apps - nothing specific to Intel or AMD. THe Phenom I have is SSE 1 - 4a si that should not be the issue. I am choosing the 64 bit installer and I redownloaded the installer last night as well.

I assume you had the SSE2 app chosen last time, this time try the SSE app.

Claggy

I will try it with SSE but then I should just run the basic app as according to the installer - there are no optimizations below SSE2.

---

What is MB7_win_x86_SSE2_CPU_r1717_nonIPP_FFTW.exe found here, http://mikesworldnet.de/produkte.html
Has anyone tried that App on one of these AMD CPUs that doesn't seem to like the other MB7 CPU Apps?
Just wondering.

Installed with SSE and forced an update:

3/2/2014 10:34:04 AM | | Reading preferences override file
3/2/2014 10:34:04 AM | | Preferences:
3/2/2014 10:34:04 AM | | max memory usage when active: 4095.59MB
3/2/2014 10:34:04 AM | | max memory usage when idle: 2047.79MB
3/2/2014 10:34:04 AM | | max disk usage: 1.00GB
3/2/2014 10:34:04 AM | | suspend work if non-BOINC CPU load exceeds 25%
3/2/2014 10:34:04 AM | | (to change preferences, visit a project web site or select Preferences in the Manager)
3/2/2014 10:34:04 AM | | Not using a proxy
3/2/2014 10:34:05 AM | SETI@home | Restarting task 27mr13ac.11577.23380.438086664201.12.10_1 using setiathome_v7 version 700 (cuda50) in slot 0
3/2/2014 10:34:05 AM | SETI@home | Starting task 27mr13ac.11577.23380.438086664201.12.230_1 using setiathome_v7 version 700 (cuda50) in slot 1
3/2/2014 10:34:05 AM | SETI@home | Sending scheduler request: To fetch work.
3/2/2014 10:34:05 AM | SETI@home | Requesting new tasks for CPU and NVIDIA
3/2/2014 10:34:07 AM | SETI@home | Scheduler request completed: got 27 new tasks
3/2/2014 10:34:09 AM | SETI@home | Started download of 24se13ad.15412.14893.438086664197.12.13

As far as I can tell - only GPU units. Probably because of the 1 unit enforcement.

---

As far as I can tell - only GPU units. Probably because of the 1 unit enforcement.

Another option is to revert to Stock, you'll get CPU Wu's since you won't be using the SETI@home v7 (anonymous platform, CPU) planclass with it's 1 task per day,
But you'd also get GPU work from a mixture of Cuda planclasses until one settles down to be fastest, and that might not be the real fastest app,
but there are ways around that, although it might takes a few days to weeks to achieve it.

Claggy

I guess I will have to go back to the regular app. I reset the project and now my CUDAs are erroring with the same error as the CPU work.

---

So far, after reverting to stock - the first 4 v7 7.00 cpu work has errors. One astropulse CUDA and several 7 7.00 CUDA work finished ok...

---

You could try running the app outside of BOINC in standalone mode. This way you might get better error message than the "too many exit(0)s".

How to run stand-alone?

---

How to run stand-alone?

I was going to ask that as I only run SETI. It ran fine before BOINC.

---

How to run stand-alone?

Make a folder somewhere.
Copy setiathome_7.00_windows_86_64.exe , libfftw3f-3-3_upx.dll and one workunit (filenames might be a bit off) from projects\setiathome.berkeley.edu to the folder.
Rename the workunit to work_unit.sah .
Open command prompt and cd to the folder.
Run setiathome_7.00_windows_86_64.exe from command line.

Post the filenames, whatever output setiathome_7.00_windows_86_64.exe gives and contents of stderr.txt.

cuda42 and cuda50 seems to be failing on your machine as well. Someone else needs to tell what files those use.

It has finally gotten better the last 2 weeks or so without running work outside of BOINC. Now doing Astropulse 6.01 on the CPU instances.

---