SETI@home Version 7 has been released

Message boards : News : SETI@home Version 7 has been released
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 7 · 8 · 9 · 10 · 11 · 12 · 13 . . . 20 · Next

AuthorMessage
Profile Gone with the wind Crowdfunding Project Donor*Special Project $75 donor
Volunteer tester

Send message
Joined: 19 Nov 00
Posts: 41574
Credit: 41,972,691
RAC: 355
Message 1374994 - Posted: 1 Jun 2013, 22:19:44 UTC

Not whatever. www.berkely.edu and setiathome.berkely.edu do not exist. You're missing an 'e' in the address and the 'l' is in another place. www.berkeley.edu and setiathome.berkeley.edu will get you places.

I understand what you are saying, but you have to appreciate that many people here don't have English as their first language. And a lot of those that do aren't that good at it either. This is an International project!

ID: 1374994 · Report as offensive
Profile Bogie
Volunteer tester
Avatar

Send message
Joined: 21 Dec 06
Posts: 84
Credit: 69,492,395
RAC: 0
United States
Message 1375000 - Posted: 1 Jun 2013, 22:28:28 UTC

Gatekeeper i am really sorry i am just not understandin it all ,


Make a new text file named app_config.xml and have this in the contents

<app_config>
<app>
<name>setiathome_v7</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>.04</cpu_usage>
</gpu_versions>
</app>
</app_config>

add that after i run outa work, maybe a sticky with all the steps for someone like that isnt catching, again i am sorry
ID: 1375000 · Report as offensive
Profile Ageless
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 14848
Credit: 4,057,196
RAC: 1,064
Netherlands
Message 1375006 - Posted: 1 Jun 2013, 22:38:02 UTC - in response to Message 1374994.  
Last modified: 1 Jun 2013, 22:38:39 UTC

Not whatever. www.berkely.edu and setiathome.berkely.edu do not exist. You're missing an 'e' in the address and the 'l' is in another place. www.berkeley.edu and setiathome.berkeley.edu will get you places.

I understand what you are saying, but you have to appreciate that many people here don't have English as their first language. And a lot of those that do aren't that good at it either. This is an International project!

Going to the right address has in this case nothing to do with the language people speak, type, read or breathe. So, when this person is constantly speaking of how berkely.edu was not available to him, then that's correct. It's not available to me either, or to you, as it doesn't exist. Has nothing to do with their (in)ability to use a different language, which in the case of JR Love is English (since he's American).

So, wrong tree you're barking at.
Jord

According to Giorgo of the Ancient Astronaut Theorists I do not help with tech questions via private message. He's right: please use the forums for that.
ID: 1375006 · Report as offensive
James W Project Donor

Send message
Joined: 26 May 12
Posts: 47
Credit: 3,912,276
RAC: 5,084
United States
Message 1375012 - Posted: 1 Jun 2013, 22:46:10 UTC - in response to Message 1374937.  

For the rest of the SETI users still using stock applications, they haven't noticed a thing as evidenced by not a single one complaining in here. Only those using the Anonymous platform are having issues, compounded by those that still can't tell the difference between BOINC and SETI@home.


Well let me be the "first stock user" to offer a complaint. Thursday I got a set of various v7 Cuda wu which all errored out because the v7 dll (or whatever)were stalled in downloading. When I updated, I got all the remaining V7 files and the errored wu's were sent back. I got another set of cuda wu's, but these had "computing errors" because of "too many errors." The only v7 wu I have at present is a cuda42 being processed by my GPU.

Now my BOINC manager is requesting work about every 20-25 minutes, and getting the responses:
06/01/2013 3:33:04 PM | SETI@home | Sending scheduler request: To fetch work.
06/01/2013 3:33:04 PM | SETI@home | Not requesting tasks
06/01/2013 3:33:06 PM | SETI@home | Scheduler request completed
06/01/2013 3:38:11 PM | SETI@home | Sending scheduler request: To fetch work.
06/01/2013 3:38:11 PM | SETI@home | Not requesting tasks
06/01/2013 3:38:13 PM | SETI@home | Scheduler request completed

ETC.

I'm now running my backup project (Rosetta) for my CPU until I can start getting SETI wu's again. From one of Eric's comments, I gather that when wu's error out, the scheduler won't sent that type of work again for a period of time?
ID: 1375012 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 11480
Credit: 167,606,403
RAC: 106,528
Australia
Message 1375013 - Posted: 1 Jun 2013, 22:46:37 UTC - in response to Message 1374987.  
Last modified: 1 Jun 2013, 22:51:30 UTC

I'm sure by now most of you realise that it's pointless trying to help this person (JR Love) as they need professional help that we are unable to provide?
Grant
Darwin NT
ID: 1375013 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 11480
Credit: 167,606,403
RAC: 106,528
Australia
Message 1375016 - Posted: 1 Jun 2013, 22:50:35 UTC - in response to Message 1375012.  

Thursday I got a set of various v7 Cuda wu which all errored out because the v7 dll (or whatever)were stalled in downloading.

You are the first person to have this issue. I and others that have posted in the appropriate forums all had WUs download & ready to crunch, and they sat there & waitined untill the other files had finished downloading, which in my case took most of the night. Then they started processing.
But the fact is they didn't error out while waiting for the application files to downlond.
The ones that errored out on my other system did so becasue of my fiddling with the system.
Grant
Darwin NT
ID: 1375016 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 13034
Credit: 143,094,868
RAC: 196,136
United Kingdom
Message 1375032 - Posted: 1 Jun 2013, 23:16:17 UTC - in response to Message 1375016.  

Thursday I got a set of various v7 Cuda wu which all errored out because the v7 dll (or whatever)were stalled in downloading.

You are the first person to have this issue. I and others that have posted in the appropriate forums all had WUs download & ready to crunch, and they sat there & waitined untill the other files had finished downloading, which in my case took most of the night. Then they started processing.
But the fact is they didn't error out while waiting for the application files to downlond.
The ones that errored out on my other system did so becasue of my fiddling with the system.

I think there have been one or two other reports (buried under all the noise) of tasks being run, and failing, before a necessary program or DLL has completed downloading.

Of course, the most likely explanation would be a user-cancelled transfer or the client calling a permanent transfer failure (which itself warrants investigation), but there have been just enough (I think) plausible reports to take seriously.

This would be a BOINC client bug, of course, not SETI's responsibility, but if anybody can document an occurrence (in Number Crunching, of course), we know how to take it further.
ID: 1375032 · Report as offensive
Eric Korpela Project Donor
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 3 Apr 99
Posts: 1344
Credit: 44,331,725
RAC: 58,611
United States
Message 1375042 - Posted: 1 Jun 2013, 23:23:47 UTC - in response to Message 1375012.  

06/01/2013 3:33:04 PM | SETI@home | Sending scheduler request: To fetch work.
06/01/2013 3:33:04 PM | SETI@home | Not requesting tasks
06/01/2013 3:33:06 PM | SETI@home | Scheduler request completed


That's either a bug in BOINC, or some really confusing messages from BOINC about what it's attempting to do.
@SETIEric

ID: 1375042 · Report as offensive
Philip Bott

Send message
Joined: 16 Sep 99
Posts: 3
Credit: 71,704,615
RAC: 25,257
Australia
Message 1375050 - Posted: 1 Jun 2013, 23:32:34 UTC - in response to Message 1373503.  

Hi Raistmer,
I'm no expert at software, but I'd like to have a go at helping with testing. Version 7 is working fine on the 4 core MacPro. ie 2 x 2.66 GHz Dual Core Intel Xeon with 16 GB 667 Mhz DDR2 FB-DIMM , with Graphics NVIDIA GeForce 8800 GT 512 MB . It can't go past Mac OSX 10.7.5

The dual six core, 12 processor MacPro is "Running" 12 work units [w/u] after installing Version 7 and now giving a "Waiting for shared memory " messages beside all the other ready to start w/u. Previously it was hyper-threading 24 w/u simultaneously. It appeared that hyper-threading was less efficient as the time taken was 5+ hours for SETI@home v6 and 45++ for Astro-Pulse v6. I tried to exclude Astro-Pulse in the preferences but they still come through occasionally. Currently the 12 core is doing average SETI@home v7 w/u in around 3 hours and AstroPulse around 20 hours. The "time remaining" for AstroPulse is dropping from around 35hrs to under 25 hrs as the unit is processed towards completion. Please advise any machine details and other information required to assist.

The 12 core MacPro is dedicated only for SETI and has minimal other software installed beside OSX 10.8.3 (12D78) and Firefox. It has a non-standard processor set, ie. 2 x 3.33 GHz 6-core Intel Xeon when I purchased it. Since SETI didn't use much memory in my experience, I dropped the installed memory to 2 of 8 GB 1333 MHz DDR3 blocks ie 16 GB total at present. Graphic Card is ATI Radeon HD 5870 1024 MB. When running full-time both machines pushed my RAC temporarily under 400, but my electricity bill jumped alarmingly, so the 12core is on off-peak discount tariff currently and only running about half the time, ie late at night and over the weekends.


ID: 1375050 · Report as offensive
Profile cormac
Avatar

Send message
Joined: 22 Nov 05
Posts: 4
Credit: 6,677,386
RAC: 0
United States
Message 1375103 - Posted: 2 Jun 2013, 3:25:18 UTC - in response to Message 1373233.  

hi jason

saw you mentioned in another post in this area as a guru for the lunatics ... i've gotten the .xml file corrected so seti (and i also use enstien as a backup) is running again ....

i was wondering if there was anyway i could tell that the lunatic overclocking setting were still in place? forgive my ignorance but i am just not well-versed on the inside workings of this program

thanks for any help :)
take care,
cormac :)

ID: 1375103 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1375105 - Posted: 2 Jun 2013, 4:04:48 UTC - in response to Message 1374870.  


Only errors: hostid=6598140.

What is here the problem?


Could be a bad/overheating card or bad driver. My bigger concern is why the daily quotas are being reduced when the invalid results are found. That needs to be fixed.

Please do Not 'fix' this.
Computers that spit invalid results have to be punished somehow.
Else they will do that for 1000s of tasks per day.

Bigger issue is when 2 bad results (wrong overflow) match each other (so are marked valid) and the real good result is marked as invalid.


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1375105 · Report as offensive
Profile betreger
Avatar

Send message
Joined: 29 Jun 99
Posts: 9345
Credit: 24,674,934
RAC: 21,765
United States
Message 1375112 - Posted: 2 Jun 2013, 4:54:01 UTC - in response to Message 1375105.  

Not the way I see it.
Computers that spit invalid results have to be punished somehow.

One should not punish someone who is trying to do the right thing and failing, one should just try to minimize the damage and try to help them improve. This is part of the cost of doing business, the way it is done in distributed computing.
ID: 1375112 · Report as offensive
Profile betreger
Avatar

Send message
Joined: 29 Jun 99
Posts: 9345
Credit: 24,674,934
RAC: 21,765
United States
Message 1375116 - Posted: 2 Jun 2013, 5:18:50 UTC - in response to Message 1375105.  

Computers that spit invalid results have to be punished somehow.

In fact on second thought punishment would be counterproductive to the project.
ID: 1375116 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3720
Credit: 9,385,827
RAC: 0
Bulgaria
Message 1375117 - Posted: 2 Jun 2013, 5:23:36 UTC - in response to Message 1375112.  

Not the way I see it.
Computers that spit invalid results have to be punished somehow.

One should not punish someone ...

I didn't say "punish someone" (human, user), I say punish "Computers that spit invalid results"
(and by 'punish' I mean in this case to only reduce the tasks those Computers can have per day until the Computer is fixed, not let those Computers burn through 1000s of tasks and generate invalids)


 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 
ID: 1375117 · Report as offensive
Profile Sutaru Tsureku
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7104
Credit: 147,313,424
RAC: 0
Germany
Message 1375118 - Posted: 2 Jun 2013, 5:24:38 UTC - in response to Message 1373842.  
Last modified: 2 Jun 2013, 5:28:54 UTC

The first invalid result of my machine (The WU was calculated correct! I'm a beta tester and the app is running very well on my system.), because of two wingman with 'SETI@home v7 v7.03 (opencl_ati5_sah)' apps and wrong results. :-(
So this WU is uncrunched and lost?

26mr10ab.22208.11282.8.12.10 - wuid=1255961084

OK, OK, the result have 5x 0.
But the next WU have maybe 'better' findings?!

Both wingmen are running too low an APP runtime version, need to be an APP SDK 2.6 driver/APP runtime :

Driver version: CAL 1.4.1607 (VM)
Version: OpenCL 1.1 AMD-APP-SDK-v2.5 (793.1)

Claggy


I found this hostid=6343230.

This machine D/L 8 MB AP files just for to trash them immediately.
Wasted bandwidth.

This machine have also a too old driver installed?

Why send SAH then WUs to such hosts...
Maybe the admins could adjust something for to prevent this?


* Best regards! :-) * Philip J. Fry, team seti.international founder. * Optimize your PC for higher RAC. * SETI@home needs your help. *
ID: 1375118 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 11480
Credit: 167,606,403
RAC: 106,528
Australia
Message 1375119 - Posted: 2 Jun 2013, 5:26:18 UTC - in response to Message 1375116.  
Last modified: 2 Jun 2013, 5:26:46 UTC

Computers that spit invalid results have to be punished somehow.

In fact on second thought punishment would be counterproductive to the project.

What would be counter productive is allowing invalid results in to the science database. This is a science project, the accuracy & integrity of the data is important.
If you're going to trash Wurk units, you don't get any more until you are able to return valid work.

If someone has a problem, they can ask for help in the appropriate forums. If they don't ask, it's ovbiously not important to them. But it is important to the project so their ability to trash work is limited.
Grant
Darwin NT
ID: 1375119 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 11480
Credit: 167,606,403
RAC: 106,528
Australia
Message 1375120 - Posted: 2 Jun 2013, 5:27:57 UTC - in response to Message 1375112.  

One should not punish someone who is trying to do the right thing and failing, one should just try to minimize the damage and try to help them improve.

Hence the limit on the amount of damage they can do until they fix the problem.

Grant
Darwin NT
ID: 1375120 · Report as offensive
Eric Korpela Project Donor
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 3 Apr 99
Posts: 1344
Credit: 44,331,725
RAC: 58,611
United States
Message 1375122 - Posted: 2 Jun 2013, 5:37:22 UTC - in response to Message 1375105.  
Last modified: 2 Jun 2013, 5:37:58 UTC


Could be a bad/overheating card or bad driver. My bigger concern is why the daily quotas are being reduced when the invalid results are found. That needs to be fixed.

Please do Not 'fix' this.
Computers that spit invalid results have to be punished somehow.
Else they will do that for 1000s of tasks per day.


Sorry, typo, that should have said "aren't being reduced". Despite all the bad results that computer is still getting 100 results per day per app version.
@SETIEric

ID: 1375122 · Report as offensive
Profile Jeff Buck Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 11 Feb 00
Posts: 1441
Credit: 148,764,870
RAC: 0
United States
Message 1375123 - Posted: 2 Jun 2013, 5:38:28 UTC - in response to Message 1375012.  


Now my BOINC manager is requesting work about every 20-25 minutes, and getting the responses:
06/01/2013 3:33:04 PM | SETI@home | Sending scheduler request: To fetch work.
06/01/2013 3:33:04 PM | SETI@home | Not requesting tasks
06/01/2013 3:33:06 PM | SETI@home | Scheduler request completed
06/01/2013 3:38:11 PM | SETI@home | Sending scheduler request: To fetch work.
06/01/2013 3:38:11 PM | SETI@home | Not requesting tasks
06/01/2013 3:38:13 PM | SETI@home | Scheduler request completed

ETC.


I had this happen on each of my GPU machines as soon as each one started processing its first v7 WU. BOINC reported all completed tasks, then went into the above loop with about a 5 minute communications backoff after each. After a couple hours on the first machine, I tried the manual "Update" project button which seemed to force a work request and cleared the loop. Unfortunately, that technique didn't work on the other three machines. The manual Update request did force a work request, but 5 minutes later the "fetch work / Not requesting tasks" loop resumed. I finally got it cleared by using the "Read config file" command on the "Advanced" menu. I don't actually have a config file, but it seemed to force a reset of the defaults and everything returned to normal from then on.

By the way, this problem did not occur on my CPU-only machines. It seems to be something triggered by the first v7 GPU task that a machine runs.
ID: 1375123 · Report as offensive
Eric Korpela Project Donor
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 3 Apr 99
Posts: 1344
Credit: 44,331,725
RAC: 58,611
United States
Message 1375124 - Posted: 2 Jun 2013, 5:40:07 UTC - in response to Message 1375118.  

Maybe the admins could adjust something for to prevent this?


The change to prevent this went in yesterday. Let me know if its still happening.
@SETIEric

ID: 1375124 · Report as offensive
Previous · 1 . . . 7 · 8 · 9 · 10 · 11 · 12 · 13 . . . 20 · Next

Message boards : News : SETI@home Version 7 has been released


 
©2019 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.