I\'ve been running BOINC since January on my OS X G5 with the boinc 4.19 client. I\'m concerned that within the results window I\'ve got outcomes of \"client errors\", \"unknown\", etc...

Is my processing time adding any value and are my results useful at all?

Thanks,
Cliff

if you get the three phase graphs that means that you went through OK. there seems to be a bug with my zip library on macs after going through the hundreds of files to zip up. but by that point the upload files were built & sent, the last "megatrickle" (i.e. for phase graph info) was sent etc, so the 'real science output' is complete.

Unfortunately we just don't have the hardware (we develop and test on an older G4 laptop) to really do the Mac, and there's only two of us computer guys on the project, so we're stretched out a bit thin. hopefully we are getting some money for a dualproc G5 which should help our development greatly (the laptop takes 5-6 months to do a hadsm3 run!)

> if you get the three phase graphs that means that you went through OK. there
> seems to be a bug with my zip library on macs after going through the hundreds
> of files to zip up. but by that point the upload files were built & sent,
> the last "megatrickle" (i.e. for phase graph info) was sent etc, so the 'real
> science output' is complete.
>
> Unfortunately we just don't have the hardware (we develop and test on an older
> G4 laptop) to really do the Mac, and there's only two of us computer guys on
> the project, so we're stretched out a bit thin. hopefully we are getting some
> money for a dualproc G5 which should help our development greatly (the laptop
> takes 5-6 months to do a hadsm3 run!)
>
>

Thanks for your reply. On the bottom of the results window, I get 3 phase plots each for Temp and precipitation. Is that correct?

yeah, that means you have completed a run & uploaded fine. the graphs "trickle" and uploading starts before the big process of zipping up every file. Our problem is that on our old G4 laptop (our only Apple Mac hardware -- hopefully not for long) we have never made it to the end of a run! It's the same code as on linux, but for some reason after 300 files or so people get this "zip" error.

If you can do me a favor --- can you see how many files are .zip and how many aren't? I wonder how far this actually happens. I may have to go to using another compression at least for doing the 300 files (maybe gzip/zlib).

> if you get the three phase graphs that means that you went through OK. there
> seems to be a bug with my zip library on macs after going through the hundreds
> of files to zip up. but by that point the upload files were built & sent,
> the last "megatrickle" (i.e. for phase graph info) was sent etc, so the 'real
> science output' is complete.

To be more space-efficient, maybe I could zip them up myself or at least mvoe them to external/media storage. I was poking poking around the /projects/climateprediction.net//datain and dataout directory and it seems that there is a very logical structure. Can you please reccommend what can be zipped ?

Thanks,
Cliff

> Can you please reccommend what can be zipped ?

If you hadn't experienced the zip problem you would have ended up with the following files moved to your 33t9_200166369 directory:

33t9_200166369.xml (the uncompressed file already in that directory)
All of the result files (dataout/33t9*.zip)
dataout/restart.day.zip
dataout/yabsd.out.zip
datain/phase2.start.zip
datain/phase3.start.zip

All other files and the working subdirectories are deleted, and you can backup the directory and recover BOINC free space by deleting it.
<br><a href="http://www.teampicard.net/"><img src="http://www.teampicard.net/images/picardmini.gif"></a><a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/team_display.php?teamid=3">Join us here</a>