Message boards : Number crunching : work units??
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 . . . 6 · Next

AuthorMessage
Profile [AF>Futura Sciences>Linux] Thr...

Send message
Joined: 6 Mar 07
Posts: 8
Credit: 31,454
RAC: 0
Message 17844 - Posted: 13 Sep 2007, 11:07:16 UTC

Waiting ^^

Thancks for your precision.
------
Thrr-Gilag Kee'rr

L'Alliance Francophone
ID: 17844 · Report as offensive     Reply Quote
Profile [SG-2ndWave] Joey

Send message
Joined: 3 Jan 06
Posts: 1
Credit: 13,970
RAC: 0
Message 17845 - Posted: 13 Sep 2007, 16:00:49 UTC

All WUs have gone and I only got 15 of them. :-(
ID: 17845 · Report as offensive     Reply Quote
Profile Neasan
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 30 Nov 06
Posts: 234
Credit: 11,078
RAC: 0
Message 17846 - Posted: 13 Sep 2007, 16:02:53 UTC

I'm running this show and got none, I think 15 is quite good to be honest we do have a lot of volunteers who will have gotten nothing close to 15 if they got any.

Anyway I hope everyone else is happy?
ID: 17846 · Report as offensive     Reply Quote
Profile cthulhu2007

Send message
Joined: 20 Jan 07
Posts: 13
Credit: 4,099
RAC: 0
Message 17847 - Posted: 13 Sep 2007, 16:03:19 UTC

Hot dawg!!!

I just checked my BOINC client and found to my delight that I have not 1 not 2 but THREE WU's! BOINC is currently crunching one of them and the other 2 are standing by (as well as SETI and EINSTEIN)... There was a comment on another thread about folks stuck with dialup not being able to get WU's because folks with broadband were hogging all the work but that is patently not true. I'm on dialup and I got work.
ID: 17847 · Report as offensive     Reply Quote
watnou

Send message
Joined: 1 Sep 04
Posts: 101
Credit: 1,395,204
RAC: 0
Message 17848 - Posted: 13 Sep 2007, 16:17:04 UTC

you just have to be lucky i guess
only 1 machine (out of 4) was able to get some wu's
/me happy
ID: 17848 · Report as offensive     Reply Quote
Tjeerd

Send message
Joined: 26 Aug 07
Posts: 1
Credit: 10,468
RAC: 0
Message 17849 - Posted: 13 Sep 2007, 16:21:33 UTC

I just saw that I got a whopping 68 WU's :0
ID: 17849 · Report as offensive     Reply Quote
larry1186

Send message
Joined: 4 Oct 06
Posts: 38
Credit: 24,908
RAC: 0
Message 17850 - Posted: 13 Sep 2007, 16:41:27 UTC
Last modified: 13 Sep 2007, 16:43:08 UTC

Wahoo!! I got 11 WUs!!

Unfortunately, every last one of them suffered the same fate...


9/13/2007 9:51:30 AM|lhcathome|Deferring communication for 7 sec
9/13/2007 9:51:30 AM|lhcathome|Reason: requested by project
9/13/2007 9:53:49 AM||Project communication failed: attempting access to reference site
9/13/2007 9:53:49 AM|lhcathome|Backing off 1 min 0 sec on download of file logo_machine_light_1.01_.tga
9/13/2007 9:53:51 AM||Access to reference site succeeded - project servers may be temporarily down.
9/13/2007 9:57:20 AM|lhcathome|[error] Signature verification failed for logo_machine_light_1.01_.tga
9/13/2007 9:57:20 AM|lhcathome|[error] Checksum or signature error for logo_machine_light_1.01_.tga
9/13/2007 9:57:21 AM|lhcathome|Deferring communication for 1 min 0 sec
9/13/2007 9:57:21 AM|lhcathome|Reason: Unrecoverable error for result wi11m11101ne_i11m11101ne__44__64.28_59.31__8_10__5__10_1_sixvf_boinc293703_2
(app_version download error: couldn't get input files:<file_xfer_error> <file_name>logo_machine_light_1.01_.tga</file_name> <error_code>-120</error_code> <error_message>signature verification failed</error_message></file_xfer_error>)
9/13/2007 9:57:21 AM|lhcathome|Deferring communication for 2 min 12 sec
9/13/2007 9:57:21 AM|lhcathome|Reason: Unrecoverable error for result wi11m11101ne_i11m11101ne__44__64.28_59.31__8_10__5__50_1_sixvf_boinc293707_4 [...]
9/13/2007 9:57:21 AM|lhcathome|Deferring communication for 4 min 10 sec
9/13/2007 9:57:21 AM|lhcathome|Reason: Unrecoverable error for result wi11m11101ne_i11m11101ne__45__64.28_59.31__12_14__5__20_1_sixvf_boinc293776_4 [...]
9/13/2007 9:57:21 AM|lhcathome|Deferring communication for 23 min 51 sec
9/13/2007 9:57:21 AM|lhcathome|Reason: Unrecoverable error for result wi11m11101ne_i11m11101ne__45__64.28_59.31__12_14__5__60_1_sixvf_boinc293780_2 [...]
9/13/2007 9:57:21 AM|lhcathome|Deferring communication for 1 hr 39 min 22 sec
9/13/2007 9:57:21 AM|lhcathome|Reason: Unrecoverable error for result wi11m11101ne_i11m11101ne__45__64.28_59.31__8_10__5__60_1_sixvf_boinc293764_4 [...]
9/13/2007 9:57:21 AM|lhcathome|Deferring communication for 1 min 0 sec
9/13/2007 9:57:21 AM|lhcathome|Reason: Unrecoverable error for result wi11m11101ne_i11m11101ne__47__64.28_59.31__2_4__5__30_1_sixvf_boinc293849_0 [...]
9/13/2007 9:58:26 AM|lhcathome|Fetching scheduler list
9/13/2007 9:58:31 AM|lhcathome|Master file download succeeded
9/13/2007 9:58:42 AM|lhcathome|Deferring communication for 1 min 0 sec
9/13/2007 9:58:42 AM|lhcathome|Reason: no work from project

{the [...] is representative of the same error description/code as previous}
Oh well, at least I had a chance...
ID: 17850 · Report as offensive     Reply Quote
Ravens

Send message
Joined: 14 Dec 06
Posts: 29
Credit: 128,225
RAC: 0
Message 17851 - Posted: 13 Sep 2007, 16:44:43 UTC

I leave my system always online, but missed all the WU's - again. It's depressing.
System looks for work, finds none, so progressively sets longer intervals before looking again. Gets to up to 2 1/2 hours, these long intervals pretty well ensure I will miss any avail batches of data.
I've got network connect set for 4 days, LHC at 47% resource use, 1.4GHz Pentium, run Rosetta mostly while waiting.
Any ideas on what to change to shorten those long intervals bewteen connections to LHC? Thanks - Mike
ID: 17851 · Report as offensive     Reply Quote
Profile [AF>Futura Sciences>Linux] Thr...

Send message
Joined: 6 Mar 07
Posts: 8
Credit: 31,454
RAC: 0
Message 17852 - Posted: 13 Sep 2007, 16:48:36 UTC - in response to Message 17848.  

you just have to be lucky i guess
only 1 machine (out of 4) was able to get some wu's
/me happy

Be happy, none of my five computers got any of them :sad:

Let some to me next time :)
------
Thrr-Gilag Kee'rr

L'Alliance Francophone
ID: 17852 · Report as offensive     Reply Quote
zombie67 [MM]
Avatar

Send message
Joined: 24 Nov 06
Posts: 76
Credit: 7,953,478
RAC: 5
Message 17853 - Posted: 13 Sep 2007, 16:51:36 UTC

I have 11 machines attached, and received no WUs.
Dublin, California
Team: SETI.USA

ID: 17853 · Report as offensive     Reply Quote
daemion

Send message
Joined: 4 Dec 06
Posts: 33
Credit: 75,491
RAC: 0
Message 17854 - Posted: 13 Sep 2007, 17:51:20 UTC - in response to Message 17851.  
Last modified: 13 Sep 2007, 18:47:11 UTC

I leave my system always online, but missed all the WU's - again. It's depressing.
System looks for work, finds none, so progressively sets longer intervals before looking again. Gets to up to 2 1/2 hours, these long intervals pretty well ensure I will miss any avail batches of data.
I've got network connect set for 4 days, LHC at 47% resource use, 1.4GHz Pentium, run Rosetta mostly while waiting.
Any ideas on what to change to shorten those long intervals bewteen connections to LHC? Thanks - Mike


I don't believe there is a way to prevent BOINC from increasing the time it checks the server, its designed to prevent the various project's servers from being bombarded with pointless requests. i have found that by setting your network connect to something really small (i usually have it set to 0.01) then it checks it more regularly, especially when you are about to finish the WU you are currently on. Then when there actually are WU i temporarily bump my cache to 3 days lets say, then when the WUs are gone reset it to an uber low number so they upload as the finish. this way i usually get about one day of WU for each of my computers as BOINC NEVER gives me the correct amount of work my settings indicate. (eg. even when its set to 0.001 i still get 6 hours of WUs from the project.)

people will probably hate me for this but thats how i got about 85 WU across my 5 comps, though one did not get any. it also helps that im a student who is usually hanging around his computer during the day :P

:edited for the dreaded SP:
ID: 17854 · Report as offensive     Reply Quote
KAMasud

Send message
Joined: 7 Oct 06
Posts: 114
Credit: 23,192
RAC: 0
Message 17855 - Posted: 13 Sep 2007, 17:56:43 UTC


:( waited three days and got non???????????? and i am connected 24/7? :(
ID: 17855 · Report as offensive     Reply Quote
Profile caspr
Avatar

Send message
Joined: 26 Apr 06
Posts: 89
Credit: 309,235
RAC: 0
Message 17856 - Posted: 13 Sep 2007, 18:14:40 UTC
Last modified: 13 Sep 2007, 18:15:28 UTC

@ daeimon... how do u bump up u'r cache?
I'm lost when it comes to that.
A clear conscience is usually the sign of a bad memory


ID: 17856 · Report as offensive     Reply Quote
daemion

Send message
Joined: 4 Dec 06
Posts: 33
Credit: 75,491
RAC: 0
Message 17858 - Posted: 13 Sep 2007, 18:45:54 UTC - in response to Message 17856.  

@ daeimon... how do u bump up u'r cache?
I'm lost when it comes to that.


Its pretty straightforward, go to "your account" page, then click on "general preferences." now you will need to know which preference setting you have your computers set to (--- or default, home, work, school. you can check this by looking at your individual host(s) from the "computers on this account" section of the "your account" page.) then simply modify the preference settings pertaining to the host you are concerned with under "connect to network about every" XXX days where XXX is the size of the cache you desire.

a word of warning, if you crunch for projects that give you "extra" stats for being the first one to complete the WU (such as ABC) then a large cache won't work out well (as you will probably only confirm someone else's results.) on the other hand if you are attached to a project with stability issues (read nearly all projects) then a small cache might cause your PC to run out of WU for the duration of the outage (assuming of course you don't run more than one project.)

HTH
ID: 17858 · Report as offensive     Reply Quote
Profile caspr
Avatar

Send message
Joined: 26 Apr 06
Posts: 89
Credit: 309,235
RAC: 0
Message 17892 - Posted: 14 Sep 2007, 15:21:14 UTC

Its pretty straightforward, go to "your account" page, then click on "general preferences." now you will need to know which preference setting you have your computers set to (--- or default, home, work, school. you can check this by looking at your individual host(s) from the "computers on this account" section of the "your account" page.) then simply modify the preference settings pertaining to the host you are concerned with under "connect to network about every" XXX days where XXX is the size of the cache you desire.

a word of warning, if you crunch for projects that give you "extra" stats for being the first one to complete the WU (such as ABC) then a large cache won't work out well (as you will probably only confirm someone else's results.) on the other hand if you are attached to a project with stability issues (read nearly all projects) then a small cache might cause your PC to run out of WU for the duration of the outage (assuming of course you don't run more than one project.)



@daemion... thanx! I see what u'r talkin about now. I'll give it a shot next time some WU's come up!
A clear conscience is usually the sign of a bad memory


ID: 17892 · Report as offensive     Reply Quote
PovAddict
Avatar

Send message
Joined: 14 Jul 05
Posts: 275
Credit: 49,291
RAC: 0
Message 17932 - Posted: 20 Sep 2007, 0:23:57 UTC - in response to Message 17931.  
Last modified: 20 Sep 2007, 0:24:20 UTC

can anyone say "hammer the server"?
I can say it; but it's a very bad idea.

Can anyone tweak the parameters mentioned in this thread.
Only the server admins can. It's server-side parameters.
ID: 17932 · Report as offensive     Reply Quote
PovAddict
Avatar

Send message
Joined: 14 Jul 05
Posts: 275
Credit: 49,291
RAC: 0
Message 17934 - Posted: 20 Sep 2007, 2:36:59 UTC - in response to Message 17933.  
Last modified: 20 Sep 2007, 2:38:32 UTC

Aye, it will make the admins unhappy. But if crunchers don't hammer the server they won't get work and they'll be unhappy. And if only a few crunchers get work (like now) then LHC will have to have IR=5 and then I will be unhappy. It's better for admins to be unhappy than crunchers be unhappy.

Users unhappy -> users quitting the project -> less users asking for work -> more work for the ones who stay. But by telling everybody how to hammer the server, it would be the same as if nobody did it, with the difference that server overload may make it even more difficult to get work.

Only the server admins can. It's server-side parameters.

Exactly. Server side.

I think the admins should set min_sendwork_interval to an insane value like 3600 to make your stupid "solution" stop working.
ID: 17934 · Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 . . . 6 · Next

Message boards : Number crunching : work units??


©2024 CERN