I started up my Climateprediction work as usual when I fired up my PC (SUSE Linux 10.0) this morning, and got the following messages:

2006-04-20 11:32:03 [---] May run out of work in 0.20 days; requesting more
2006-04-20 11:32:03 [climateprediction.net] Requesting 8922 seconds of work
2006-04-20 11:32:03 [climateprediction.net] Sending request to scheduler: http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2006-04-20 11:32:04 [climateprediction.net] Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
2006-04-20 11:32:04 [climateprediction.net] Message from server: No work sent (won\'t finish in time) Computer available 37.0% of time, this project gets 100.0% of that
2006-04-20 11:32:04 [climateprediction.net] Message from server: No work sent (won\'t finish in time) Computer available 37.0% of time, this project gets 100.0% of that
2006-04-20 11:32:04 [climateprediction.net] No work from project
2006-04-20 11:32:04 [climateprediction.net] No work from project
2006-04-20 11:32:04 [climateprediction.net] Deferring communication with project for 1 hours, 0 minutes, and 0 seconds
2006-04-20 11:32:04 [climateprediction.net] Deferring communication with project for 1 hours, 0 minutes, and 0 seconds

I had a look at my current Result ID (816403) and saw:
Sent: 3 May 2005
Time reported or deadline: 16 April 2006
Server state: Over
Outcomee: No reply
Client state: New
CPU time: 1,256,944.00

Is this telling me:
(a) I\'ve reached a deadline date (16 April 2006) and Climateprediction has given up waiting to receive the results of the work unit that\'s been running on my PC for a year; and
(b) I\'ve wasted my time (and my PC\'s time)?

I hope not - I really want to help this project.

---

Looking at the messages, in particular: Outcome: No reply, I think that it may be
c) It\'s given up waiting to hear from your model.
Taking over a year is OK, but your model MUST trickle at least every couple of weeks to keep it \'alive\'.

Anyway, it was a slab model, and they became obsolete late last year.
The project has moved on to the next part of the experiment, with Coupled Ocean models, TCM.
These models don\'t leave behind any data the way slab and sulphur did, and they are a single phase of 160 years.

And they require BOINC version 5.* to function. The current latest version is 5.2.13
With the one that you have, it will be necessary to uninstall it first, and then install the latest version.

Hmm, I believe my model has been trickling every 3-4 days recently.

So, this slab model I\'ve been running is no use any more. I have been wasting my time.

Should I now take the following action to continue:
(a) halt the execution of the current slab model and delete its data;
(b) uninstall the version of BOINC I currently have;
(c) install version 5.2.13; and
(d) download a new model and start on it.

Have I missed any messages telling me my slab model is obsolete, and my BOINC is out of date?

---

I think that your computer is a bit slow for the project, but if you stick at it, you\'ll get there. Although you may only have time for one TCM model.

Yes, steps a-d seem reasonable.

The only place that \'messages\' about the different models has appeared are on the boards, mainly on the php \'Message Board\', but it\'s been mentioned in replys to people on thsese help boards. The TCMs started February 14.

There used to be a \"Message from the server: a new version of BOINC is available for download\", but I can\'t remember if it still appears, as I\'ve had 5.2.8 since late last year.

The reason for the need to uninstall, is that after about 4.2? (can\'t remember), the design of BOINC changed completely, so that now it\'s split into parts, the gui, and a worker, (daemon, client).
And an even more different version is due soon. And has been for a few months. :(

If you cancel your model, you may be interested in SAP http://attribution.cpdn.org/ ?! The SAP model needs much less CPU time than the CPDN model (but that project is limited to 10000 runs and will be closed then).

---

Linux Users Everywhere @ BOINC

If you cancel your model, you may be interested in SAP http://attribution.cpdn.org/ ?! The SAP model needs much less CPU time than the CPDN model (but that project is limited to 10000 runs and will be closed then).

Hi, Desti,

My guess is that his machine\'s CPU and memory limitations could cause problems on Seasonal Model. Even it the Model runs, it would surely interfere with any other work on the machine. (Recommended minimum P4 2.4 or equivalent and 1GB of memory.)

---

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.

Thanks for your responses guys. I\'m stopping my Work Unit and deleting BOINC.

Best wishes for the project.

---

Thanks for your responses guys. I\'m stopping my Work Unit and deleting BOINC.

Best wishes for the project.

Hello,

that is very unfortunate, is there no other project (with much less system requirements), which sounds interesting for you?