Message boards :
Number crunching :
How to run the test with reference WU?
Message board moderation
Author | Message |
---|---|
![]() ![]() Send message Joined: 1 Jul 02 Posts: 139 Credit: 1,466,611 RAC: 0 ![]() |
Hi, I'm trying to execute the reference WU with the last official SETI@home client (5.15). I've created a folder with the reference WU that is in the source zip (reference_work_unit.sah -> work_unit.sah), SETI@home 5.15 client and fftw dll (libfftw3f-3-1-1a.dll). When I start the client, it not start to crunch, but give a stderr.txt with this error: SETI@home error -108 Unknown error from boinc_init_diagnostics() File: ../main.cpp Line: 227 Why? I'm doing something wrong or it's a bug of the new client? With the old client there was not this problem I remember... Please let me know the right way, bye. GHz BOINC.Italy |
![]() ![]() Send message Joined: 9 Jul 99 Posts: 1199 Credit: 6,615,780 RAC: 0 ![]() |
You need a file called init_data.xml in the same directory. Also, you can run the client with -standalone as an option. Right-click the link below, and choose "save". http://www.zadra.org/seti_enhanced/downloads/init_data.xml HTH, Simon. Donate to SETI@Home via PayPal! Optimized SETI@Home apps + Information |
![]() ![]() Send message Joined: 1 Jul 02 Posts: 139 Credit: 1,466,611 RAC: 0 ![]() |
You need a file called init_data.xml in the same directory. Thanks very much Simon! Using you init_data.xml it start to run! It works also with a minimal data into, like this: http://www.webalice.it/landipaolo/file/init_data.xml It run OK, but it write somthing in the stderr.txt file. At start this line: Can't set up shared mem: -1 And some second after this: ar=0.775000 NumCfft=58347 NumGauss= 304962714 NumPulse= 58444196736 NumTriplet= 4979562790912 It's normal or not? GHz BOINC.Italy |
![]() ![]() Send message Joined: 22 Mar 04 Posts: 442 Credit: 459,063 RAC: 0 ![]() |
You need a file called init_data.xml in the same directory. Hi Simon, Further to this, how does one check the validity of the result.How do we check for "strongly similiar" results. Does it have something to do with running "rescmp.exe" ? Thanks. |
Astro ![]() Send message Joined: 16 Apr 02 Posts: 8026 Credit: 600,015 RAC: 0 |
It run OK, but it write somthing in the stderr.txt file. At start this line: I got the same error when running the wu that failed for DR. ![]() simon said "Tony, that always happens when you run S@H apps on their own (not through BOINC). So it's not a real error " |
![]() ![]() Send message Joined: 9 Jul 99 Posts: 1199 Credit: 6,615,780 RAC: 0 ![]() |
Yes, the message about "cannot set up shared mem" is normal - it always gets printed out when you do offline testing (i.e. not in BOINC). The other line is a statistics output and is also normal (check out all your results online, you'll see that line is there if you're using the standard science app). Validating results offline is done with rescmp, yes. If you want, you can just download a test pack from my site that has all these things and scripts already included :o) To successfully validate your results with rescmp, you will have to grab the reference WU and reference results, as well - rescmp needs to have something to compare your result to. Then, when you have them, do "rescmp reference_result.sah result.sah" or whatever you named the files. HTH, Simon. Donate to SETI@Home via PayPal! Optimized SETI@Home apps + Information |
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 ![]() |
It run OK, but it write somthing in the stderr.txt file. At start this line: To explain further: When there is an init_data.xml file the S@H app tries to establish the shared memory interface to BOINC. The simple fact that the app was started by the user rather than started by BOINC is the cause of that message; it's not an error though it looks like one. Joe |
Idefix Send message Joined: 7 Sep 99 Posts: 154 Credit: 482,193 RAC: 0 ![]() |
Hi, I've another question concerning the standalone testing: I've just tested various applications (stock, Crunch3r's, Simon's ...) with the init_data.xml and work_unit.sah of Simon's "CPU Test + Benchmark Tool". But how do you meassure the cpu time? I read in another thread that you will find the final cpu time in the <wu_cpu_time> line of the init_data.xml file. But this line rather seems to be the time of the last checkpoint. At the moment I'm comparing the creation time and last modified time of the stderr.txt file to get the cpu time. But is there any better method? Regards, Carsten |
©2023 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.