Im very new to this CPDN thing but i have a query. My results page shows i have 259.2 (or something like that credits, but when i look i have 2 trickles recorded. Any idea why i don\'t have credit for both of them or is there some sort of delay/lag in the result being credited. Im just a bit curious

---

The system calculate the credits at least once a day. There were issues when it was real-time, so they went to a batch system. It was running twice a day (every 12 hours) at one point, but currently I believe it is now only once a day (from what I have observed).

---

Thanks its nice to know i was worried a WU might have gone wrong or something. Its also credited now so it must have just run.

---

I must have a problem then, CPDN hasn\'t givenme any credit since Dec 27th. That is the day I started up my new 3.2 Ghz P4 box. I started with a clean drive and down loaded all new WU\'s but I still have not gotten and credit from CPDN. Who/where do I go to ask for help to fix this.

---

Most of your recent models have had \'aborted by user\' against them. If this is deliberate then I\'d suggest you have a look at the \'no more work\' button whieh prevents surplus models from being downloaded if the PC is busy.

If you want to temporarily disable Boinc while playing a game, then open Boinc, Activity menu, \'suspend\'. DON\'T use the suspend button on the work/tasks tab because this does something else.

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/results.php?hostid=516677

A number also have \'exit code 0\' listed. This can happen when Windows kills the task, which happens via Task Manager, \'kill process\', and also if the PC shuts down too quickly. Right-clicking and selecting \'exit\' on the icon just before you shut down is a good idea to avoid this.

Finally, I would recommend backing up Boinc at intervals. Take a look at the various README posts here for more information:

http://www.climateprediction.net/board/viewforum.php?f=36

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

I had been having problems with my old motherboarda and the 1.9 Ghz P4 chip that was in it. Then I lost my primary hard drive. I got a new 80 GB Hard Drive, a new INTEL P4 Socket 478 Mobo at ZipZoomFly and a 3.2 Ghz P4E CPU chip. When I restarted my box up and recommectedto BONIC CPDN sent me 3 files one right after the other. I knew that there was no way that I would finish 3 files in time so I aborted 2 of them. I am getting ready to take on 2-3 more BONIC projects and I wanted to be sure all files get processed and on time. Still doesn\'t explain why I haven\'t gotten any credit for 2 weeks.

---

Your computer keeps crashing the models.
Look at your Results page here.

Also, your computer hasn\'t uploaded any trickles since the 13th of December, so there is no reason for the server to give you credit.

Check in the folders of the models to see if there are any files labelled: trickle_up_...............
If there are some, then you probably have a firewall problem.

Well, since january 9th i haven\'t got a single credit - although my machine is nice trickling away. Anything wrong with the granting mechanisms for credits?

I\'m not worried for the credits but for the science values...

---

greetz, Uli

I\'m getting mine, both here and at BBC.

Sometimes a model can start \'looping\' if they get to an impossible climate (only the 5.08 ones), or more commonly there can be firewall issues which block the model talking to the servers.

Is there anything in your \'messages\' tab in the Boinc manager which looks odd? (if in doubt, just paste the lot here).

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

There is nothing unusual in the message tab ( i know, what i\'m talking about ;) ). The only unusual thing on that day was, it was the original deadline. Since then, the new deadline is january 8th 2008.

This is the result in question:
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5480755

---

greetz, Uli

There is nothing unusual in the message tab ( i know, what i\'m talking about ;) ). The only unusual thing on that day was, it was the original deadline. Since then, the new deadline is january 8th 2008.

This is the result in question:
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5480755

Could your model be looping? While it is trickling, the link you provided doesn\'t show any new trickles since the 8th. If your model is looping and sending the same trickle over and over it won;t get credited.

Just my $.02 worth.

---

Mark

Maybe I\'m missing something but where I can se number of trickles?
I have two WU running. One for 24 hours, other for 8 hours. (1.06% and 0.45% of WU). If I calculated corectly I should report at leas two tricles. But my credit is stil at 0.
Any help?

thnx.

btw. WU are:
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/workunit.php?wuid=5882146
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/workunit.php?wuid=5828210

Tom

1st trickle is at about 0.67%
They occur roughly on 4th December, starting at the end of the 1st FULL year. You may see the data being collected on the 3rd, if you happen to be watching.

Once the trickle is on the server, you then have to wait until the credit program runs, which is once per day now, because of the huge load on the servers.

The trickles will appear on the page for each of your models.
This is one of them.
You can get there in many ways, one of which is by clicking on your name to the left of here, then on \"view\" to the right of Computers, then on the number underneath Results, then clicking on the one that you want to look at in the \"Result ID\" column.

Finally, if you DO look at your name to the left of here, you\'ll see that you now have some credit.

There is nothing unusual in the message tab ( i know, what i\'m talking about ;) ). The only unusual thing on that day was, it was the original deadline. Since then, the new deadline is january 8th 2008.

This is the result in question:
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=5480755

Could your model be looping? While it is trickling, the link you provided doesn\'t show any new trickles since the 8th. If your model is looping and sending the same trickle over and over it won;t get credited.

Just my $.02 worth.

It might be a good idea to post the historical log (stdoutdae.txt) for the 7th + 8th of Jan since there should have been messages in the log about the deadline. A looper might be indicated by zero status exits, and in addition to that, Boinc would have been going into panic mode due to the deadline (possibly followed by a silent period due to LTD?).

I guess the PC is offline for a lot of the time looking at the trickle history (i.e., the batch of trickles uploaded on the 2nd)?

What version of Boinc was it?

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

I\'m trying to answer all questions:

The Computer was cut off from the net over the holidays and was crunching away. On january 2nd it uploaded the big bunch of trickles piled up over the holidays.

The BOINC version is 5.7.5.

Just now it trickled again, here is the log:

...
17.01.2007 13:06:08|climateprediction.net|Sending scheduler request: To send trickle-up message
17.01.2007 13:06:08|climateprediction.net|(not requesting new work or reporting completed tasks)
17.01.2007 13:06:27|climateprediction.net|Scheduler RPC succeeded [server version 507]
...

Maybe i have really a looping model, just found the following in the long term log:

...
2007-01-10 22:01:19 [climateprediction.net] Resuming task hadcm3lbm_4f8c_05162505_1 using hadcm3lb version 508
2007-01-10 23:44:10 [climateprediction.net] Task hadcm3lbm_4f8c_05162505_1 exited with zero status but no \'finished\' file
2007-01-10 23:44:10 [climateprediction.net] If this happens repeatedly you may need to reset the project.
2007-01-10 23:44:10 [climateprediction.net] Restarting task hadcm3lbm_4f8c_05162505_1 using hadcm3lb version 508
2007-01-11 00:04:49 [climateprediction.net] Task hadcm3lbm_4f8c_05162505_1 exited with zero status but no \'finished\' file
2007-01-11 00:04:49 [climateprediction.net] If this happens repeatedly you may need to reset the project.
2007-01-11 00:04:49 [climateprediction.net] Restarting task hadcm3lbm_4f8c_05162505_1 using hadcm3lb version 508
...

Just saw it was going back from >78% to 77,757% done. :(
I think, i should abort this unit with >1800 hrs work done?

---

greetz, Uli

I\'d let it cycle through once more, just to make absolutely sure, but yes, it does look like you\'re going to have to kill it.

Basically the model has reached a point where it\'s climate is unrealistic, so it tries to rerun the most recent year just in case it was due to a quirky floating point calculation. The older code (5.08) kept cycling around, whereas the newer 5.15 code automatically aborts itself after it\'s retried the year.

If it\'s looping, the model has produced a lot of scientific information which has uploaded to the servers, and your 1800 hours has produced dividends, but there\'s no point in it continuing further if the model has started to drift out of reality.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

I\'d let it cycle through once more, just to make absolutely sure, but yes, it does look like you\'re going to have to kill it.
(...)
If it\'s looping, the model has produced a lot of scientific information which has uploaded to the servers, and your 1800 hours has produced dividends, but there\'s no point in it continuing further if the model has started to drift out of reality.

I had a closer look into the long term log. The model cycled at least 7 times through the same year since 8th january. So i decided to abort the unit and immediately got a new one.

Thanks everybody for the helpful posts. :D

---

greetz, Uli