Help - cruncher does lots of work for no return

Message boards : Number crunching : Help - cruncher does lots of work for no return
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1061360 - Posted: 30 Dec 2010, 10:39:58 UTC

One of my crunchers periodically goes into a sort of spasm and does lots of effort, but no progress is made on one or more WU. I generally let these tasks run until the "time to go" is zero, with the progress also zero!

The cruncher in question (5562708) is a quad core AMD with 3Gb RAM, running Ubuntu and is generally a stable beast, its last reboot followed a security update, but this problem existed before this update arrived.

Suggestions please.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1061360 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1061384 - Posted: 30 Dec 2010, 12:15:04 UTC - in response to Message 1061360.  

Had a look at your LINUX (Ubuntu)host, all errors are -177 errors, explained in the BOINC Wiki Faq.

ERR_RSC_LIMIT_EXCEEDED -177

Maximum disk usage is exceeded. The science application is aborted.
Check your disk usage settings in your general preferences.

Maximum disk usage exceeded or Maximum memory usage exceeded (check the messages tab for more information) The science application is aborted. Check your disk/memory usage settings in your general preferences.

Maximum CPU time exceeded. The maximum granted CPU time for a work unit is exceeded, probably a stuck work unit. The science application is aborted. If a work unit gets stuck, exiting and restarting BOINC is a known work around to fix this (if it's not already aborted ;-)).

You may want to look here.

Shouldn't this be in Questions and Answer forum?


ID: 1061384 · Report as offensive
Profile Gundolf Jahn

Send message
Joined: 19 Sep 00
Posts: 3184
Credit: 446,358
RAC: 0
Germany
Message 1061432 - Posted: 30 Dec 2010, 14:19:48 UTC - in response to Message 1061360.  

See also this thread.

Gruß,
Gundolf
ID: 1061432 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1061452 - Posted: 30 Dec 2010, 15:18:45 UTC - in response to Message 1061384.  

Had a look at your LINUX (Ubuntu)host, all errors are -177 errors, explained in the BOINC Wiki Faq.

ERR_RSC_LIMIT_EXCEEDED -177

Maximum disk usage is exceeded. The science application is aborted.
Check your disk usage settings in your general preferences.

Maximum disk usage exceeded or Maximum memory usage exceeded (check the messages tab for more information) The science application is aborted. Check your disk/memory usage settings in your general preferences.

Maximum CPU time exceeded. The maximum granted CPU time for a work unit is exceeded, probably a stuck work unit. The science application is aborted. If a work unit gets stuck, exiting and restarting BOINC is a known work around to fix this (if it's not already aborted ;-)).

You may want to look here.

Shouldn't this be in Questions and Answer forum?


It certainly does not look like a -177, as the total available to BOINC projects is 93Gb, and only 3.8Gb are used. Other WU run quite happily concurrently with these errant WU.

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1061452 · Report as offensive
Profile Gundolf Jahn

Send message
Joined: 19 Sep 00
Posts: 3184
Credit: 446,358
RAC: 0
Germany
Message 1061500 - Posted: 30 Dec 2010, 16:43:19 UTC - in response to Message 1061452.  

It certainly does not look like a -177, as the total available to BOINC projects is 93Gb, and only 3.8Gb are used. Other WU run quite happily concurrently with these errant WU.

It certainly does look like a -177, you just didn't read all paragraphs. The relveant one is number three:
Maximum CPU time exceeded. The maximum granted CPU time for a work unit is exceeded, probably a stuck work unit. The science application is aborted. If a work unit gets stuck, exiting and restarting BOINC is a known work around to fix this (if it's not already aborted ;-)).

Your tasks are aborted with "maximum CPU time exceeded", because they (presumably) got stuck in the initialisation phase.

Did you read the thread I linked to?

Gruß,
Gundolf
Computer sind nicht alles im Leben. (Kleiner Scherz)

SETI@home classic workunits 3,758
SETI@home classic CPU time 66,520 hours
ID: 1061500 · Report as offensive
Profile gizbar
Avatar

Send message
Joined: 7 Jan 01
Posts: 586
Credit: 21,087,774
RAC: 0
United Kingdom
Message 1061555 - Posted: 30 Dec 2010, 18:40:10 UTC
Last modified: 30 Dec 2010, 18:40:39 UTC

Just for info, and not having checked the links provided at all, AMD CPU's seem to have a problem starting work units using the stock application. Not sure if it applies to both Windows and Linux, but I'm guessing it does.

One way around it is to install the Lunatics optimised app, as that doesn't lock up like the stock app.

Only problem is that I don't know if you can install Lunatics on Linux.

Giz.


A proud GPU User Server Donor!
ID: 1061555 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1061560 - Posted: 30 Dec 2010, 19:00:48 UTC - in response to Message 1061555.  

it's more of a hit and miss per computer. Some AMD CPU's run the stock app fine. Others crash and burn on it.
I'd recommend the optimized apps for any seti AMD CPU


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1061560 · Report as offensive
-BeNt-
Avatar

Send message
Joined: 17 Oct 99
Posts: 1234
Credit: 10,116,112
RAC: 0
United States
Message 1061561 - Posted: 30 Dec 2010, 19:03:19 UTC

Running stock or lunatics seems to run perfectly with my FX-60. Of course like previously stated I guess it varies computer to computer.
Traveling through space at ~67,000mph!
ID: 1061561 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1061658 - Posted: 30 Dec 2010, 23:13:52 UTC - in response to Message 1061555.  

Just for info, and not having checked the links provided at all, AMD CPU's seem to have a problem starting work units using the stock application. Not sure if it applies to both Windows and Linux, but I'm guessing it does.

One way around it is to install the Lunatics optimised app, as that doesn't lock up like the stock app.

Only problem is that I don't know if you can install Lunatics on Linux.

Giz.

Its strange, I thought I spotted the Linux MB downloads when I wasn't looking for them. Now I'm looking for them they appear to have gone into hiding - I can only see Astropulse for Linux on the download page....
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1061658 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1061732 - Posted: 31 Dec 2010, 1:13:04 UTC - in response to Message 1061658.  

try downloading from herehttp://www.arkayn.us/seti/


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1061732 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1061852 - Posted: 31 Dec 2010, 9:53:47 UTC

Thanks,
A question, I'm not sure if the processor is "SSE2" or "SSE3" - I assume that SSE3 is backwards compatible with SSE2, but not the other way round, so if I use the SSE2 app it will work either way. But if the processor is SSE2 and I load the SSE3 app it will not work. Or is there a way of finding which processor model I've got "from the keyboard"?
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1061852 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1061854 - Posted: 31 Dec 2010, 10:03:41 UTC - in response to Message 1061852.  

Answered part of my own question - its an "AMD Athlon(tm) II X4 620 Processor", now to find that on the AMD website and see what its capable of.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1061854 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14680
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1061857 - Posted: 31 Dec 2010, 10:17:53 UTC - in response to Message 1061852.  

Thanks,
A question, I'm not sure if the processor is "SSE2" or "SSE3" - I assume that SSE3 is backwards compatible with SSE2, but not the other way round, so if I use the SSE2 app it will work either way. But if the processor is SSE2 and I load the SSE3 app it will not work. Or is there a way of finding which processor model I've got "from the keyboard"?

Much the easiest is to download and run CPU-Z.

Your assumption is correct - you can run the SSE2 app on both SSE2 and SSE3 chips, but you cannot run the SSE3 app on a SSE2-only chip.
ID: 1061857 · Report as offensive
-BeNt-
Avatar

Send message
Joined: 17 Oct 99
Posts: 1234
Credit: 10,116,112
RAC: 0
United States
Message 1061890 - Posted: 31 Dec 2010, 11:45:42 UTC - in response to Message 1061854.  

Answered part of my own question - its an "AMD Athlon(tm) II X4 620 Processor", now to find that on the AMD website and see what its capable of.


This site has a bunch of information on different processors. Great all around site for information. It says your processor supports these instruction sets:

MMX
3DNow!
SSE
SSE2
SSE3
SSE4a
Advanced Bit Manipulation
AMD64 technology
AMD-V (virtualization) technology
Enhanced Virus Protection

So install the SSE3 apps. Hope it helps good luck!
Traveling through space at ~67,000mph!
ID: 1061890 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1061906 - Posted: 31 Dec 2010, 12:58:51 UTC

Many thanks.
I'd not had a chance to look at the AMD site thanks to the phone going off too much, and only me here to answer it (with increasing impatience to the glazing and insulation cold callers....)
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1061906 · Report as offensive
Metod, S56RKO
Volunteer tester

Send message
Joined: 27 Sep 02
Posts: 309
Credit: 113,221,277
RAC: 9
Slovenia
Message 1062036 - Posted: 31 Dec 2010, 19:04:47 UTC - in response to Message 1061852.  
Last modified: 31 Dec 2010, 19:07:35 UTC

A question, I'm not sure if the processor is "SSE2" or "SSE3" - I assume that SSE3 is backwards compatible with SSE2, but not the other way round, so if I use the SSE2 app it will work either way. But if the processor is SSE2 and I load the SSE3 app it will not work. Or is there a way of finding which processor model I've got "from the keyboard"?


You already got your answer ... I just want to include this one for completeness. You said you're running linux, so here's proper Linux command:

cat /proc/cpuinfo


which gives whole lotta information about every logical CPU (physical core as well as HT siblings):

processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 65
model name      : Dual-Core AMD Opteron(tm) Processor 2218
stepping        : 2
cpu MHz         : 2600.000
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 2
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow rep_good extd_apicid pni cx16 lahf_lm cmp_legacy svm extapic cr8_legacy
bogomips        : 5200.00
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp tm stc


The flags section gives your information. In my case it lists MMX, MMXext, 3Dnow, 3DnowExt, SSE and SSE2.
Metod ...
ID: 1062036 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 1062052 - Posted: 31 Dec 2010, 19:41:24 UTC - in response to Message 1062036.  

A question, I'm not sure if the processor is "SSE2" or "SSE3" - I assume that SSE3 is backwards compatible with SSE2, but not the other way round, so if I use the SSE2 app it will work either way. But if the processor is SSE2 and I load the SSE3 app it will not work. Or is there a way of finding which processor model I've got "from the keyboard"?


You already got your answer ... I just want to include this one for completeness. You said you're running linux, so here's proper Linux command:

cat /proc/cpuinfo


which gives whole lotta information about every logical CPU (physical core as well as HT siblings):
...
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow rep_good extd_apicid pni cx16 lahf_lm cmp_legacy svm extapic cr8_legacy
...

The flags section gives your information. In my case it lists MMX, MMXext, 3Dnow, 3DnowExt, SSE and SSE2.

It also lists pni, an abbreviation for Prescott New Instructions, and that's the original terminology for SSE3.
                                                              Joe
ID: 1062052 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1062080 - Posted: 31 Dec 2010, 21:09:01 UTC - in response to Message 1062052.  

And it does it for each of the four processors in the proverbial blink of an eye.


I'm going to set No New Tasks and empty my chache before doing anything too exciting, just so the it starts with a clean sheet.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1062080 · Report as offensive
Metod, S56RKO
Volunteer tester

Send message
Joined: 27 Sep 02
Posts: 309
Credit: 113,221,277
RAC: 9
Slovenia
Message 1062682 - Posted: 2 Jan 2011, 11:23:50 UTC - in response to Message 1062052.  

It also lists pni, an abbreviation for Prescott New Instructions, and that's the original terminology for SSE3.


Oh, my, I never noticed that one ... I was sure I had a SSE2 CPU. Gotta get a SSE3 science application.
Metod ...
ID: 1062682 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 22572
Credit: 416,307,556
RAC: 380
United Kingdom
Message 1064743 - Posted: 8 Jan 2011, 20:44:31 UTC

While waiting for the errant Linux box to finish its cache I decided to install the Lunatics optimised app on one of my windoze machines. What a difference, the GPU (geforce 210) is getting through WU in between 30 and 35 minutes instead of a couple of hours or more. The CPU however doesn't look to be much better, maybe an hour or two off its previous 16.
Time will tell what this will do to my RAC, but its got to get better - my first guess is that the GPU on its own will be giving me about 4000, compared with the three cruncher total of 4000 before the Lunatics arrived.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1064743 · Report as offensive
1 · 2 · Next

Message boards : Number crunching : Help - cruncher does lots of work for no return


 
©2025 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.