Posts by Khangollo

61) Message boards : Number crunching : Panic Mode On (74) Server problems? (Message 1232644)
Posted 15 May 2012 by Profile Khangollo
Post:
All downloads that were stuck on my hosts now went "through":
[error] File 22ap11ad.25473.17253.14.10.23 has wrong size: expected 375365, got 0
[error] Checksum or signature error for 22ap11ad.25473.17253.14.10.23

That's a new one :)
Now I have more errors than some doofus with broken and forgotten GPU.

Edit: nevermind. fresh WUs are downloading fine again
62) Message boards : Number crunching : Almost time for the next outage isn't it ??? (Message 1232608)
Posted 15 May 2012 by Profile Khangollo
Post:
Here's your outage now!
63) Message boards : Number crunching : Four more Astropulses on server monitor (Message 1231426)
Posted 13 May 2012 by Profile Khangollo
Post:
Looks like one already got stuck again :-)
64) Message boards : Number crunching : WU count (Message 1230529)
Posted 11 May 2012 by Profile Khangollo
Post:
Probably not. 2147483647 is the tricky number (upper limit of a 32-bit signed integer), so we still have 1147483647 to go.
65) Message boards : Number crunching : APv6 workunits are getting stuck in validation, too (Message 1230506)
Posted 11 May 2012 by Profile Khangollo
Post:
So, after ~12.000 Astropulse v505 tasks finally got cleared, it seems that APv6 is also affected by the same bug!

Are project admins even aware of this annoying bug? Is it going to get fixed?

Example: http://setiathome.berkeley.edu/workunit.php?wuid=949723426
66) Message boards : Technical News : Feijoada (Apr 30 2012) (Message 1226370)
Posted 2 May 2012 by Profile Khangollo
Post:
Old-school??? This phone is just some new-age gizmo.
Real phones have rotary dials.

If everyone still had those, there would be a lot less unnecessary calls.
67) Message boards : Number crunching : Scheduler request failed: Failure when receiving data from the peer (Message 1226351)
Posted 2 May 2012 by Profile Khangollo
Post: 
<cc_config>
  <options>
    <max_tasks_reported>150</max_tasks_reported>
  </options>
</cc_config>

This will limit reporting to 150 tasks at a time.

Put cc_config.xml in your BOINC's data directory - that's at the top level (where files like client_state.xml are also located) and then click on "Advanced / Read config file" menu in Boinc Manager.
68) Message boards : Number crunching : Boinc 7.0.24/Ubuntu computation errors (Message 1226281)
Posted 2 May 2012 by Profile Khangollo
Post: 
ps -auxf

Will give you a complete process tree. Use Shift+Pg Up/Pg Down to scroll up and down the console.
Or, if you want to filter it to just some processes with "boinc" name in it,
ps -auxf | grep boinc

(or just ps -axf if the output is too wide; will not show the user accounts under which processes run)
69) Message boards : Number crunching : AstroPulse v6.02: 16.59 credits for 91.7 hours' processing?! (Message 1225293)
Posted 30 Apr 2012 by Profile Khangollo
Post:
This always happens when you're running anonymous platform and your wingman is running nVidia/ATI Astropulse GPU application (CreditNew - the scoring system s@h uses doesn't like it at all).
Weird enough, you're running stock application, so this shouldn't happen. But it did.
My suggestion is don't run astropulse on slow computers. Stupid CreditNew will "rip you off" often, so at least let it rip you off on shorter workunits.
70) Message boards : Number crunching : Panic Mode On (74) Server problems? (Message 1225093)
Posted 29 Apr 2012 by Profile Khangollo
Post:
Ok.
MB assimilators seem to be offline for a while now. Hopefully we won't have a legit panic in a day or two... right through holidays.
71) Message boards : Number crunching : Boinc 7.0.24/Ubuntu computation errors (Message 1225087)
Posted 29 Apr 2012 by Profile Khangollo
Post:
I haven't used an init.d for boinc, I always just fired it up manually- the server is on 24-7 and doesn't get rebooted often if ever. But that's another option to try.

In that case, you can always just run it from your home dir (~/BOINC) and download the 6.12.34 from Berkeley - or better yet - 6.10.60 version for now. That one works 100%. Wait with packaged v7 until it gets sorted out.
72) Message boards : Number crunching : Boinc 7.0.24/Ubuntu computation errors (Message 1225073)
Posted 29 Apr 2012 by Profile Khangollo
Post:
Since the downloaded BOINC works fine why not just use that?

If it works fine, there's no point in downloading sources and compiling them.

Like I said, it's a long shot. Quite possibly, there's a bug elsewhere - boinc developers should look into that.

Does the berkely version use the /var/lib... directory or the /home/user directory?
It doesn't matter. Data directory is passed to boinc client via "--dir" command line option. If you don't specify the --dir, it uses the same directory as where the "boinc" executable is.
I don't use Debian/Ubuntu but I think the startup script, which handles starting up the boinc client as a daemon (/etc/init.d/boinc ?) should start the boinc client with appropriate --dir /var/boinc (or where the data is).
So overwriting the "boinc" executable with your own shouldn't change anything.
73) Message boards : Number crunching : Boinc 7.0.24/Ubuntu computation errors (Message 1225056)
Posted 29 Apr 2012 by Profile Khangollo
Post:
Why don't you just download the version from Berkeley, extract it your home dir, then just copy over executable ("boinc") to where the package manager's version is?

It's a long shot: if that doesn't work either, compile your own boinc from sources. This way, it will be linked exactly against libraries you have - if that is even the case for these errors. 
svn export http://boinc.berkeley.edu/svn/trunk/boinc/ boinc
cd boinc
./_autosetup
./configure --disable-server --disable-manager
make
cp client/boinc <wherever_package's_boinc_executable_is>
74) Message boards : Number crunching : Why do I get cuda workunits? (Message 1223514)
Posted 26 Apr 2012 by Profile Khangollo
Post:
You can use BoincRescheduler to reschedule existing GPU tasks to CPU.

To prevent boinc from ever doing GPU work, regardless of preferences, you can create a file cc_config.xml, put it inside BOINC's data directory (not program files). On XP, it's usually C:\Documents and Settings\All Users\Application Data\BOINC
Contents:
<cc_config>
  <options>
    <no_gpus>1</no_gpus>
  </options>
</cc_config>

(and restart BOINC).
75) Message boards : Number crunching : How much slower is Ubuntu? (Message 1222813)
Posted 24 Apr 2012 by Profile Khangollo
Post:
Current optimized applications (CPU and GPU) are the same speed.
If system is properly configured. GPU crunching can even be faster a bit and system can be run headless (no GUI, no junk services, no crap like dummy plugs and extending desktops). If you know how to do it, and is not that hard.

I really don't know where you guys come up with ridiculous figures like 2-3x.
76) Message boards : Number crunching : Inconclusive Work Units Running AP Ver 6 (Message 1222229)
Posted 23 Apr 2012 by Profile Khangollo
Post:
Special thanks to dsh for our first Linux report.

I have another - ap_6.01r546_sse3_linux64 vs. windoze stock 6.01. 30+ pulses.
( ap_24ja12ah_B4_P1_00304_20120410_15850.wu )
77) Message boards : Number crunching : Interesting factoid from Dell UK (Message 1221927)
Posted 23 Apr 2012 by Profile Khangollo
Post:
It is pretty much application-dependent. E@H applications for Linux, especially CUDA one, are significantly faster than Windows ones.

Also, I'm crunching majority of S@H work on Linux, including CUDA and am using script which raises the niceness of all CUDA-feeding processes to -10 for max. performance.
78) Message boards : Number crunching : Interesting factoid from Dell UK (Message 1221287)
Posted 21 Apr 2012 by Profile Khangollo
Post:
Distribution itself is not that relevant. It would be more useful to know, is it 3.x kernel (or some options in it) or some of the newer libraries (glibc, etc...) scientific applications use or compiler version/optimization flags they've been compiled with.
79) Message boards : Number crunching : Panic Mode On (74) Server problems? (Message 1221283)
Posted 21 Apr 2012 by Profile Khangollo
Post:
I have 11 of these. First one was in August last year, last one is from March this year (was a mistake - which I quickly corrected - to leave the 505 application in my app_info.xml because practically all remaining 505 are resends now.)
It's a shame because these WUs are probably not assimilated into science DB.
80) Message boards : Number crunching : Panic Mode On (73) Server problems? (Message 1218880)
Posted 15 Apr 2012 by Profile Khangollo
Post:
They stamped out shorties for too long,
now they got what they deserve.


Previous 20 · Next 20


 
©2024 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.