climateprediction.net home page
No Work Credit since Dec 27, 06

No Work Credit since Dec 27, 06

Questions and Answers : Windows : No Work Credit since Dec 27, 06
Message board moderation

To post messages, you must log in.

AuthorMessage
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26012 - Posted: 13 Jan 2007, 21:54:39 UTC

CPDN has not updated my work credit since Dec 27, 06, why? Re CPDN Acct # 171496. Joe B.
ID: 26012 · Report as offensive     Reply Quote
Profile MikeMarsUK
Volunteer moderator
Avatar

Send message
Joined: 13 Jan 06
Posts: 1498
Credit: 15,613,038
RAC: 0
Message 26013 - Posted: 13 Jan 2007, 22:17:58 UTC

ID: 26013 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26057 - Posted: 15 Jan 2007, 17:42:44 UTC - in response to Message 26013.  


http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=5137&nowrap=true#26006


I still have not gotten any credit, for my current work done since 12-27-06? Can anybody resolve this for me?
ID: 26057 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 26060 - Posted: 15 Jan 2007, 19:49:32 UTC

If you want credits then you\'ll have to stop aborting models and let one run.
If you still have the model sent to you in December, AND it\'s status in Tasks says: Running, then you probably have a firewall problem which is stopping trickles from getting through.
In which case there will be messages in the Messages tab saying so. (Also in the archive file: stdoutdae.txt , which is in the BOINC folder.)

Ports needed by BOINC:
BOINC.exe needs 31416 and 80 and 443 opened
BOINCmgr.exe needs 31416 opened
BOINC.scr needs 31416 opened

ID: 26060 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26061 - Posted: 15 Jan 2007, 20:36:56 UTC - in response to Message 26060.  

I do not have a firewall problem as I am still using the same one I had before I changed my Motherboard and upgraded my CPU processor from a 1.9Ghz to 3.2Ghz P4. All of my other 6 BONIC projects commuinicate correctly on a regular basis.

As I stated earlier, I only aborted those 2 files because my box downloaded them when I restarted it after the Upgrade. I didn\'t think that I could complete them (all 3) in the 6.5 months I had left along with the WUs from my other six Bonic projects.
Ref: http://www.boincsynergy.com/images/stats/comb-4156.jpg

No Problem here, if no one will help me relolve this I will just abort the present file and discontinue doing work for people who don\'t care about those trying to help their project. I\'m sure I will find another one to help within minutes. Need help by this Friday, 1-19-07, 12 PM EST.
ID: 26061 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 26062 - Posted: 15 Jan 2007, 20:43:21 UTC

We ARE trying to help. But we need some feedback about what\'s going on inside your manager/model.

What about the messages? Anything there about problems?
And what is the status for the model in Tasks.

Copy and paste of the first 20-30 lines of the messages may give us a clue.

ID: 26062 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 26064 - Posted: 15 Jan 2007, 20:49:47 UTC

Another thought:
You say that you\'re running other projects as well. Perhaps BOINC has gone into \"payback mode\", whatever it\'s called, and just won\'t run the climate model until the other projects get their fair share of the processor\'s time.

ID: 26064 · Report as offensive     Reply Quote
Profile mo.v
Volunteer moderator
Avatar

Send message
Joined: 29 Sep 04
Posts: 2363
Credit: 14,611,758
RAC: 0
Message 26080 - Posted: 16 Jan 2007, 0:58:33 UTC

Joe, the best idea would be to post about this only in this thread here.

What model dates have your 2 models reached?
Cpdn news
ID: 26080 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26085 - Posted: 16 Jan 2007, 7:00:44 UTC - in response to Message 26080.  

What model dates have your 2 models reached? First of all I only have 1 model working. File # hadcm3ohe_1dcf_05712289_1. I have used 16:53:47 CPU hours on it so far and completed .402% of the calculations. There are no adverse Msg\'s for the file. When it or I communicate with the CPDN homepage there are the usual 4 lines of communication and the last is \"Scheduler request suceeded\". I have even suspended all of the other projects for a day to let CPDN \"catch up\". Still no difference. The last comm with the home page was at 1:55:01-1:55:06, with the same result.

ID: 26085 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26086 - Posted: 16 Jan 2007, 7:04:36 UTC - in response to Message 26085.  

How do you copy the system response msgs, there is no function to do that as far as I can tell. Right clicking on a highlighted msg does nothing.
ID: 26086 · Report as offensive     Reply Quote
Profile astroWX
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1496
Credit: 95,522,203
RAC: 0
Message 26087 - Posted: 16 Jan 2007, 7:12:15 UTC
Last modified: 16 Jan 2007, 7:15:56 UTC

Do you see those two big buttons on the left of the Messages screen? They offer the choice of copying all, or selected messages.

You can select a subset of the total messages by by holding the Ctrl key and clicking away. (Standard Windows stuff.) One of the buttons copies whatever you select.

The other button copies the lot.

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.
ID: 26087 · Report as offensive     Reply Quote
Profile MikeMarsUK
Volunteer moderator
Avatar

Send message
Joined: 13 Jan 06
Posts: 1498
Credit: 15,613,038
RAC: 0
Message 26088 - Posted: 16 Jan 2007, 8:54:52 UTC


There are a number of different possibilities, and the log will help to identify the correct one:

* You have a working model, and it can\'t talk to the server for some reason

* You have a looping model (i.e., it\'s just going around in circles)

* Your model is blocked from processing for some reason (i.e., EDF, suspension, ...)

* You don\'t have a model, and Boinc doesn\'t want to download another yet (i.e., long term debt to the other projects)

We have no way to tell from \'our side\' (what we see on the website is exactly what you see when you\'re looking at someone else\'s account, i.e., not a huge amount). Hopefully when we can see what Boinc is doing in your log, we can then point the finger at something and get you moving again :-)

I'm a volunteer and my views are my own.
News and Announcements and FAQ
ID: 26088 · Report as offensive     Reply Quote
Profile mo.v
Volunteer moderator
Avatar

Send message
Joined: 29 Sep 04
Posts: 2363
Credit: 14,611,758
RAC: 0
Message 26089 - Posted: 16 Jan 2007, 14:53:10 UTC

Hang on a sec. This is Joe\'s current model

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=6256031

7 hours ago he said
I have used 16:53:47 CPU hours on it so far and completed .402% of the calculations.


Doesn\'t a model need to have completed about 0.67% to reach December 1921 and make its first trickle? Or have I misread something?

Joe, according to the trickle page of one of your previous models, your computer does about 3.8 seconds/timestep. At this crunching speed, I think your computer will need slightly more than 24 hours of processing time to complete each model year. Trickles happen on or shortly after 4 Dec of each model year.


Cpdn news
ID: 26089 · Report as offensive     Reply Quote
Profile MikeMarsUK
Volunteer moderator
Avatar

Send message
Joined: 13 Jan 06
Posts: 1498
Credit: 15,613,038
RAC: 0
Message 26091 - Posted: 16 Jan 2007, 16:13:13 UTC

That model has been running since 28th Dec and only had 16 hours of CPU time, despite having everything else suspended for a day. Something strange going on there? We really do need to see that messages log to make sense of this puzzle. Meanwhile another model has been downloaded.

Can I recommend that \'No more work\' be set against the project in the Boinc manager, this will prevent surplus models being downloaded.

Given the speed of the machine, Boinc will go into EDF (Earliest deadline first) mode if the project has less than ~40% of the total resource share, and that\'ll result in \'long term debt\' once the model is completed. I don\'t know how many projects he has, but that\'s likely to be an issue in the future.
I'm a volunteer and my views are my own.
News and Announcements and FAQ
ID: 26091 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26092 - Posted: 16 Jan 2007, 22:33:06 UTC - in response to Message 26091.  

You have to understand that I had a massive power surge because of a lighting strike very near my home the night of Dec 27. Despite a $70 surge protector, it screwed up my old Mobo and CPU chip. I ordered a new INTEL D865PERL P4 Soc 478 Mobo from Zipzoomfly.com for $107 ship incl, because I have a couple spare 3.0-3.2 Ghz P4 CPU chips. I took my box to a good local shop and they installed the new Mobo and checked the restod the system out for damage. I got the box back about 8 days later. So I didn\'t do any processing for over a week.
Then when I reloaded all of the software and restarted BONIC my system downloaded about 90 files from all of my projects. Including 3 from CPDN. Knowing that I probably could not process all 3 by the deadline I aborted 2. Last night I see that I downloaded another CPDN file. As far as speed goes, my old CPU chip was a 1.9Ghz w/ a 250Mhz FSB board, w/512Mb RD800 RAM. Now I have a 3.2Ghz P4 Extreem W/Hyperthreading & a 800Mhz FSB w/2Gb DDR3200 RAM.
Here are all my CPDN msgs from BONIC:

1/15/2007 1:42:01 AM|climateprediction.net|Restarting task hadcm3ohe_1dcf_05712289_1 using hadcm3 version 515
1/15/2007 1:42:01 AM||Allowing work fetch again.
1/15/2007 2:02:25 AM|climateprediction.net|Pausing task hadcm3ohe_1dcf_05712289_1 (removed from memory)
1/15/2007 3:02:26 AM|climateprediction.net|Restarting task hadcm3ohe_1dcf_05712289_1 using hadcm3 version 515
1/15/2007 4:37:59 AM|climateprediction.net|Pausing task hadcm3ohe_1dcf_05712289_1 (removed from memory)
1/15/2007 12:24:27 PM|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
1/15/2007 12:24:27 PM|climateprediction.net|Reason: Requested by user
1/15/2007 12:24:27 PM|climateprediction.net|(not requesting new work or reporting completed tasks)
1/15/2007 12:24:32 PM|climateprediction.net|Scheduler request succeeded
1/15/2007 8:59:19 PM||Resuming round-robin CPU scheduling.
1/15/2007 8:59:19 PM|climateprediction.net|Restarting task hadcm3ohe_1dcf_05712289_1 using hadcm3 version 515
1/15/2007 8:59:21 PM|climateprediction.net|Pausing task hadcm3ohe_1dcf_05712289_1 (removed from memory)
1/15/2007 8:59:54 PM|climateprediction.net|Restarting task hadcm3ohe_1dcf_05712289_1 using hadcm3 version 515
1/15/2007 9:00:00 PM|climateprediction.net|Pausing task hadcm3ohe_1dcf_05712289_1 (removed from memory)
1/16/2007 1:55:01 AM|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
1/16/2007 1:55:01 AM|climateprediction.net|Reason: Requested by user
1/16/2007 1:55:01 AM|climateprediction.net|(not requesting new work or reporting completed tasks)
1/16/2007 1:55:06 AM|climateprediction.net|Scheduler request succeeded
1/16/2007 1:46:32 PM|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
1/16/2007 1:46:32 PM|climateprediction.net|Reason: To send trickle-up message
1/16/2007 1:46:32 PM|climateprediction.net|(not requesting new work or reporting completed tasks)
1/16/2007 1:46:37 PM|climateprediction.net|Scheduler request succeeded

Hope this helps. Joe B.

ID: 26092 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26093 - Posted: 16 Jan 2007, 22:46:00 UTC - in response to Message 26092.  

My 1st CPDN file is at 30:00 hrs CPU time, .728% complete.
My 2nd CPDN file is at 12.22 hrs CPU time, .302% complete.
But Still No Credit showing! Joe B.

ID: 26093 · Report as offensive     Reply Quote
Profile astroWX
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1496
Credit: 95,522,203
RAC: 0
Message 26094 - Posted: 16 Jan 2007, 22:57:22 UTC
Last modified: 16 Jan 2007, 23:01:38 UTC

As noted numerous places on the Boards, trickles happen every 25920 Time Steps (one Model Year, every 0.625%). Trickles are uploaded
~04 Dec of each Model Year.

Once on the Server, they are fodder for credit. Note that credit isn\'t given instantly -- the load is heavy and the Servers have better things to do. So, there is a daily cron job. THEN, you\'ll see your credits.

Relax, it all gets sorted eventually.

You\'ve returned one Trickle, so far; credit will be forthcoming:
16 Jan 2007 18:50:46 516677 6256031 hadcm3ohe_1dcf_05712289_1 1 25,920 94,006 3.6268

[Edited for typo.]
"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.
ID: 26094 · Report as offensive     Reply Quote
Profile MikeMarsUK
Volunteer moderator
Avatar

Send message
Joined: 13 Jan 06
Posts: 1498
Credit: 15,613,038
RAC: 0
Message 26096 - Posted: 16 Jan 2007, 23:32:29 UTC

Part of the puzzle is solved here:

1/15/2007 9:00:00 PM|climateprediction.net|Pausing task hadcm3ohe_1dcf_05712289_1 (removed from memory)


Every time it removes the task from memory, you\'ll lose the progress since the last checkpoint. A checkpoint happens approx every 25 minutes on your PC, so depending on whether you\'re lucky or unlucky with preemptions and timeslices, you could be losing either not much time, or quite a bit of time.

Boinc version 5.8 and higher will be more intelligent in this regard (it\'ll only timeslice immediately after a checkpoint, so you don\'t lose time). But 5.8 is only a testing version, not yet suitable for release.

The other option is to set \'leave in memory\' to yes - but obviously only if you have sufficient system memory.

As AstroWX says, once the trickle has hit the server, you\'ll need to wait for the daily credit batch run. If I recall correctly, Ian (one of the BBC participants), said that trickles received before 3AM UK time will appear on the website approx 11AM, trickles received after 3AM will only appear the following day.
I'm a volunteer and my views are my own.
News and Announcements and FAQ
ID: 26096 · Report as offensive     Reply Quote
old_user171496

Send message
Joined: 5 Mar 06
Posts: 12
Credit: 358,032
RAC: 0
Message 26122 - Posted: 17 Jan 2007, 19:09:15 UTC - in response to Message 26096.  

Thanks, I\'ll make the change to leave the file in memory, and hit the no more work button. Appreciate the help. Joe B. Lou\'ville, KY, USA
ID: 26122 · Report as offensive     Reply Quote

Questions and Answers : Windows : No Work Credit since Dec 27, 06

©2024 climateprediction.net