First, I should say that most of the discussion on these boards is over my head, both in terms of general computing knowledge and knowledge of the CPDN project itself. So, apologies in advance for inadvertent redundancy, silly questions, wasting people’s time or screwing things up.

Here’s my problem: I set up BOINC to run climateprediction.net back in January 2005. Everything worked fine and I more or less ignored it. Then some months ago I began noticing error messages that I didn’t understand and didn’t have time to pursue. I let it go and let it go until now. But it appears that my computer hasn’t been doing any work for the project for months.

In the past couple of days I looked over some of the most recent error messages and saw a cycle of messages that went like this:

2006-08-21 21:39:25 [climateprediction.net] Unrecoverable error for result hadcm3lbm_aw5e_25277493_0 (The system cannot find the file specified. (0x2) - exit code 2 (0x2))
2006-08-21 21:39:25 [climateprediction.net] Deferring communication with project for 1 minutes and 0 seconds
2006-08-21 21:40:29 [climateprediction.net] Message from server: No work sent (reached daily quota of 1 results)
2006-08-21 21:40:29 [climateprediction.net] No work from project
2006-08-21 21:40:29 [climateprediction.net] Deferring communication with project for 19 hours, 44 minutes, and 9 seconds

followed by lots more deferrals, then the error message again (but the “result” is different, e.g., hadcm3lbm_9vuq_25230453_0), and the cycle repeats. This is pretty much what I remember seeing earlier when glancing at it.

Poked around on the message boards looking for tips but saw nothing that seemed to describe my problem. I saw stuff about loops but I don’t *think* that’s what I’m seeing. For one thing, the specific result in the error message keeps changing. I don’t really understand the stuff about loops so I can’t be sure. But I did discover that I’d been running an outdated version of BOINC (4.19). So, I decided to upgrade.

I made a backup of the BOINC folder, not using the batch file but just good old Windows copy & paste. Then I removed BOINC because somewhere I’d seen a post saying that 4.19 was a special case and had to be removed before installing newer versions. Then I downloaded and installed the new version (5.4.9).

I did not restore the backed up files. I didn’t trust that I knew what I was doing enough to risk overwriting the files in the new folder. Plus: 1) I don’t care about credits; I just want the thing to work. 2) I’m really confused about work units and knowing when something has completed anyway. 3) I figured after months of errors, and my stats showing no recent credit, the risk of interrupting ongoing work was pretty low.

Anyway, after installing the new version I started getting these messages:

9/9/2006 7:52:37 AM|climateprediction.net|Message from server: No work sent
9/9/2006 7:52:37 AM|climateprediction.net|Message from server: (reached daily quota of 1 results)

I tried updating, then re-setting the project and finally detaching from and re-attaching to the project. Again, I figured I had little to lose. None of this had any effect. I continue to get the same message.

Incidentally, at someone’s recommendation I did just now exclude the BOINC folder from Norton’s anti-virus scan. So if that was the original cause of the problem I expect it won’t happen again.

So. Any suggestions for how to get working again? Thanks.

---

Your last error message states you still have 4.19 installed, you need to us add remove utility to uninstall 4.19 and load 5.49....


<core_client_version>4.19</core_client_version>
<message>The system cannot find the file specified. (0x2) - exit code 2 (0x2)
</message>
<active_task_state>1</active_task_state>
<signal>0</signal>

---

BOINC Wiki

Thanks for the quick response. I have used the add/remove utility to uninstall BOINC and re-install 5.4.9. I still get the same error message:

9/9/2006 12:05:32 PM|climateprediction.net|Message from server: No work sent
9/9/2006 12:05:32 PM|climateprediction.net|Message from server: (reached daily quota of 1 results)

Aside from just wanting to get this working again, I have a couple questions:

Looking at my results page I see that the most recent error message is dated 9 Sep 2006 0:27:22 UTC--that would be 5:27PM Friday my time (Pacific Daylight), right? This was long before I reinstalled the newer version. Looking over the list of results, it appears that it\'s showing only one result per day. I assume that the \"daily quota of 1 result\" is the reason. Is this quota normal or did BOINC decide I was generating too many errors? Is it something I can/should (re-)set once I get this problem solved? And is it related to the \"Result duration correction factor\" listed on my computer summary page?

Again, sorry the questions are ignorant.

---

All of your crashes for the last few months are due to you not having a 5.* version of BOINC.
The new Coupled Ocean models are a \"new breed\", and upload data at shorter intervals. There are 3 types of upload, with different amounts sent for each.
The models also clean up after they finish, leaving nothing behind.
Unless they crash. (So you may have over a hundred folders of rubbish, which you can now delete.)

The models take months to complete, so when a computer starts asking the server for a new model several times a day, it\'s pretty certain that the computer is crashing models quickly.
The daily quota is to put a stop to this, before the pool of datasets is run dry.

If you wait until after your local midnight, you\'ll get another one.
But if you haven\'t fixed the problem, it too will crash.

\"Result duration correction factor\" is an attempt by each project to tell BOINC how long it will take to complete a model, relative to a \"standard\". It\'s unrelated to your problem.

Once you have a new model, it will either work, or it will crash.
If it DOES crash, the new error messages will give us a new clue, so let us know how you get on \"tomorrow\".

Thanks again for the quick reply. I\'ll wait til tomorrow and see what the new day brings.

Davis

---

Thanks again for the quick reply. I\'ll wait til tomorrow and see what the new day brings.

Davis

Actually I didn\'t have to wait til local midnight--the software sent a new request at 5:21 PM my time (just after midnight UTC) and this time the server responded. Files downloaded successfully, task is running happily! Thanks for your help Les and Jim.

Davis

---

I have this same problem and have just updated to the 5.49 and still have the same error message - do I just need to be patient and wait for tomorrow when the quota is reset or do I need to do something else?

Thanks.

Update:

28/09/2006 16:54:18|climateprediction.net|Message from server: No work sent
28/09/2006 16:54:18|climateprediction.net|Message from server: (there was work but you don\'t have enough disk space allocated)
28/09/2006 16:54:18|climateprediction.net|Message from server: Not enough disk space (only 387.2 MB free for BOINC). Review preferences for maximum disk space used.

Can anyone tell me how to fix this? I have just over 3GB of free space and can\'t find a way to allow BOINC to use more of it.

Go to Your Account, on left of this page. Find General Preferences, view or edit; then, click edit preferences. Go to Disk and memory usage section and plug-in whatever numbers you want. (I use 100 GB [more than actually available], Leave at least 0.001, Use no more than 50% of available. That works for me but you might need a higher %.) Then, click Update Preferences. Go to Projects tab in boinc manager and click Update. You\'ll be good to go.

Given that your disk is nearly full Windows could get balky. Might be a good idea to unload some infrequently-used files or folders to CD/DVD. Defrag might help a bit, too.

---

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

Thanks - it\'s now working!

---