My finger is breaking something

[Corsair]

of course my finger, not me 

I'm using optimized application for SETI and when want to reschedule I got this message in the log, and no re-schedule of course:


13 October 2010 - 20:28:05 BoincRescheduler V: 2.3


13 October 2010 - 20:28:15 Shutdown of BOINC client completed
13 October 2010 - 20:28:16 Out of range: rsc_fpops_bound < 500000000000000000.000000, total: 237
13 October 2010 - 20:28:16 Found: CPU: 0, VLAR: 0, VHAR: 0
13 October 2010 - 20:28:16 Found: GPU: 236, VLAR: 0, VHAR: 63
13 October 2010 - 20:28:16 No version for Cpu or Gpu

[fred]

of course my finger, not me  

I'm using optimized application for SETI and when want to reschedule I got this message in the log, and no re-schedule of course:


13 October 2010 - 20:28:05 BoincRescheduler V: 2.3


13 October 2010 - 20:28:15 Shutdown of BOINC client completed
13 October 2010 - 20:28:16 Out of range: rsc_fpops_bound < 500000000000000000.000000, total: 237
13 October 2010 - 20:28:16 Found: CPU: 0, VLAR: 0, VHAR: 0
13 October 2010 - 20:28:16 Found: GPU: 236, VLAR: 0, VHAR: 63
13 October 2010 - 20:28:16 No version for Cpu or Gpu

Change the following, the numbers will probably be 0.
If CPU tasks are less then 20 fill up to 20 Tasks.
And try again.

But try to reschedule no more than needed. Scheduling tasks from the CPU to GPU and back will effect the runtime prediction from the SETI server.
Meaning, next time you get work, you may momentarily see some strange runtimes.

But..... are you using SETI Lunatics optimized?

[Corsair]

1st try as follows with the set of your post (BT and BOINC Manager running):


14 October 2010 - 10:40:04 BoincRescheduler V: 2.3

14 October 2010 - 10:40:26 Found: 0 files that are lost and need to be deleted

14 October 2010 - 10:40:30 Out of range: rsc_fpops_bound < 500000000000000000.000000, total: 187
14 October 2010 - 10:40:30 Found: CPU: 0, VLAR: 0, VHAR: 0
14 October 2010 - 10:40:30 Found: GPU: 186, VLAR: 0, VHAR: 49
14 October 2010 - 10:40:30 Rescheduling needed

14 October 2010 - 10:40:37 WARNING: The BOINC Manager is running.
14 October 2010 - 10:40:37 WARNING: This may cause severe problems, as the Manager may restart the BOINC client.
14 October 2010 - 10:40:37 Shutting down BOINC client
14 October 2010 - 10:40:44 ERROR: The BOINC client is running, but it shouldn't, so something is wrong. Mutex is still set
14 October 2010 - 10:40:44 ERROR: Scan completed with an error.

[Corsair]

2nd try with BOINC manager shutdown and BT too.


14 October 2010 - 10:42:48 BoincRescheduler V: 2.3

14 October 2010 - 10:43:02 Found: 0 files that are lost and need to be deleted

14 October 2010 - 10:43:04 Out of range: rsc_fpops_bound < 500000000000000000.000000, total: 187
14 October 2010 - 10:43:04 Found: CPU: 0, VLAR: 0, VHAR: 0
14 October 2010 - 10:43:04 Found: GPU: 186, VLAR: 0, VHAR: 49
14 October 2010 - 10:43:04 Rescheduling needed

14 October 2010 - 10:43:11 Shutdown of BOINC client completed
14 October 2010 - 10:43:12 Out of range: rsc_fpops_bound < 500000000000000000.000000, total: 187
14 October 2010 - 10:43:12 Found: CPU: 0, VLAR: 0, VHAR: 0
14 October 2010 - 10:43:12 Found: GPU: 186, VLAR: 0, VHAR: 49
14 October 2010 - 10:43:13 No version for Cpu or Gpu

[Corsair]

But..... are you using SETI Lunatics optimized?

Yes I'm using them, in all computers but in the only I've nVIDIA is the one with xp pro x64.

[fred]

In the Log tab check "Debug mode".
Seti tab press "Test"
Give the log again.

[fred]

14 October 2010 - 10:40:37 WARNING: The BOINC Manager is running.
14 October 2010 - 10:40:37 WARNING: This may cause severe problems, as the Manager may restart the BOINC client.
14 October 2010 - 10:40:37 Shutting down BOINC client
14 October 2010 - 10:40:44 ERROR: The BOINC client is running, but it shouldn't, so something is wrong. Mutex is still set
14 October 2010 - 10:40:44 ERROR: Scan completed with an error.

As the warning states, best close down the Manager as it may restart or block the client from closing down.
But there are checks int the rescheduler to abort if things go wrong and that works.

[Corsair]

In the Log tab check "Debug mode".
Seti tab press "Test"
Give the log again.

here you're:

14 October 2010 - 11:50:25 BoincRescheduler V: 2.3


14 October 2010 - 11:50:56 Out of range: rsc_fpops_bound < 500000000000000000.000000, total: 185
14 October 2010 - 11:50:56 Application setiathome_enhanced found, Version: 608 , Plan class: cuda
14 October 2010 - 11:50:56 Application setiathome_enhanced found, Version: 608 , Plan class: cuda23
14 October 2010 - 11:50:56 Found: CPU: 0, VLAR: 0, VHAR: 0
14 October 2010 - 11:50:56 Found: GPU: 184, VLAR: 0, VHAR: 49
14 October 2010 - 11:50:56 Average: Ratio Gpu: 51.932532, count: 184, invalid 0
14 October 2010 - 11:50:56 Ratio CPU: min: -1.000000 max: -1.000000 - Ratio Gpu: min: 50.154052 max 54.542129
14 October 2010 - 11:50:56 Rescheduling needed

[fred]

14 October 2010 - 11:50:56 Application setiathome_enhanced found, Version: 608 , Plan class: cuda
14 October 2010 - 11:50:56 Application setiathome_enhanced found, Version: 608 , Plan class: cuda23

The Version and plan class for the CPU is missing.
That normally means no app_info.xml in C:\ProgramData\BOINC\projects\setiathome.berkeley.edu

Stop the BOINC client and copy:
If possible mail me the folder: C:\ProgramData\BOINC\projects\setiathome.berkeley.edu
And from C:\ProgramData\BOINC the file client_state.xml

[Corsair]

Fred,

I told you that my finger was doing something, off course not me    

How could be rescheduled something if is there no CPU application defined in the app_info.xml??  

now defined CPU application and reschedule working now.

[fred]

now defined CPU application and reschedule working now.

And that's why you didn't have any CPU work to begin with.
Normally the GPU is running out of work.