And I can't seem to encourage it unless I detach the project and attach a new project. I had it set to a large cache (10 days) but based on a previous post, I reset it to defaults.
BOINC is shared with climateprediction.net. CPU is a toshiba dualcore. BOINC manager is 5.8.16. Other machines on the account seem to work okay, getting new work as needed.

---

~john
N42-05 W070-40
My heart is in N56-40 E3-01

What are the time stat for the host like (the 4 near the bottom of the host summary page)?

Alinator

There's been 0 work for seti since it grabbed the initial workunit. The other 3 machines I have on the account are only running seti and seem to be doing okay.
From the summary page:
% of time BOINC client is running 89.0088 %
While BOINC running, % of time work is allowed 82 %
Average CPU efficiency 0.721967
Result duration correction factor 1.068209
And the messages from BOINC manager:
2007/08/02 08:16:50 AM|SETI@home|Sending scheduler request: Requested by user
2007/08/02 08:16:50 AM|SETI@home|(not requesting new work or reporting completed tasks)
2007/08/02 08:16:55 AM|SETI@home|Scheduler RPC succeeded [server version 511]
2007/08/02 08:16:55 AM|SETI@home|Deferring communication for 11 sec
2007/08/02 08:16:55 AM|SETI@home|Reason: requested by project

Note the "Not requesting new work" even though it does not currently have any seti workunits.

---

~john
N42-05 W070-40
My heart is in N56-40 E3-01

OK, these numbers look pretty normal for a 'daily driver' laptop, except for the CPU efficiency and the benchmarks which seem a little low, but could be due to the 'real' work you use the machine for.

So most likely the reason BOINC isn't DL'ing any work from SAH is it's seeing a deadline issue for the CPDN models currently. BOINC is pretty conservative with regards to missing deadlines and even though the deadlines seem long for CPDN, they are in fact rather 'tight' compared to the actual runtime required.

As a result it's not uncommon for BOINC to drop working on other projects for awhile when you have one onboard, especially for hosts running multiple projects and/or not run as 24/7 dedicated crunchers.

HTH,

Alinator

The clue is in the 'shared with climateprediction.net'. I take it we're talking about your Mobile Intel(R) Pentium(R) 4 CPU 3.20GHz (host 1147907 here, host 254177 at CPDN.

You have two results in progress at CPDN. One has reached 68% since 17 December 2006 (7.5 months at 4.28 sec/TS), the second hasn't yet reached 25% since 24 December (over 7 months at 4.53 sec/TS).

According to BOINC's calculations, both are probably in deadline trouble - the second one very seriously so. Also, your computer is probably a bit under-powered (446.98 MB RAM) for running two CPDN models in parallel.

Since your next CPDN trickle on the second model will be accompanied by the 40-year restart dump, now might be a good moment to consider setting 'No New Work' for CPDN and - after checking that the trickle has happened and the zip file has uploaded - aborting the second CPDN model. That's the only way you'll get any SETI work on that host.

That appears to be the answer. As soon as I set the second climatepredictor to suspend, BOINC downloaded and started a new seti workunit. It's odd that I had not seen this behavior before but it may be due to having changed the settings to "Run based on preferences" from "Run Always".
I had originally attributed this odd behavior to having set up a BOINC account manager (BOINCSTATS). When I took the machine out of the BOINC account manager, and the issue continued, I thought I'd check with others.
It all makes sense now. Thanks.

---

~john
N42-05 W070-40
My heart is in N56-40 E3-01