About three days ago (and probably in conjunction with a security upgrade forced upon my computer by Earthlink) the climate unit upon which I was working issued an error message and disappeared from my computer.

Since then I have had no luck in getting another unit to download and run. I had some overall problems with Boinc (I\'m running 5.4.11) but have resolved them...I think...Seti\'s running okay, for whatever that might mean.

At first the problem seemed to be that Boinc didn\'t know I needed a new CPDN unit. I updated, then reset the project, and when that didn\'t help, I detached, then reattached.

After the reattach, it appeared that Boinc downloaded new copies of necessary software and a work unit, but the work unit errored-out and I had to wait 24 hours before I could try for another unit.

A few minutes ago, I received the following messages:
1/16/2007 8:15:18 PM|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
1/16/2007 8:15:18 PM|climateprediction.net|Reason: To fetch work
1/16/2007 8:15:18 PM|climateprediction.net|Requesting 259200 seconds of new work
1/16/2007 8:15:20 PM|climateprediction.net|Unrecoverable error for result hadcm3ohe_1by6_05710480_1 (<file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_1.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_2.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_3.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_4.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_5.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_6.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_7.zip</file_name> <error_code>-161</error_code></file_xfer_error><file_xfer_error> <file_name>hadcm3ohe_1by6_05710480_1_8.zip</file_name> <error
1/16/2007 8:15:20 PM|climateprediction.net|Deferring scheduler requests for 1 minutes and 0 seconds
1/16/2007 8:15:24 PM|climateprediction.net|Scheduler request succeeded
1/16/2007 8:15:24 PM|climateprediction.net|Message from server: No work sent
1/16/2007 8:15:24 PM|climateprediction.net|Message from server: (reached daily quota of 1 results)
1/16/2007 8:15:24 PM|climateprediction.net|No work from project

My machine takes a long time to process a CPDN unit, but I\'d like to keep working on them. At the moment, I seem to be dead in the water.

Can you identify my problems?

Best,

Lynn Abbey

---

Actually, your machine downloaded several Models -- and crashed them.

It might help if you posted the messages under boinc Manager\'s Messages tab.

Your four most recent Models crashed before progressing far enough to create restart files. I don\'t see a reason.

What sort, specifically, did you have problems with boinc?

Perhaps you should look more closely into what that \"security upgrade\" did to your machine...? (When problems \"magically\" appear after a change, it\'s best to look into the change, eh?)

---

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

Just to add a little to AstroWX\'s answer ...

The list of your models is here :

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/results.php?hostid=489242

From the 8th of January onwards they seem to be crashing, was that when the \'security upgrade\' was applied? How much free disk space do you have?

---

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

Please forgive my ignorance, but where do I find the \"boinc Manager\'s Messages tab\"?

The problem that began occurring after the January 6 Earthlink security upgrade is that the Boinc manager program, which I run at Startup (WinXP SP2) can no longer automatically connect to the \"localhost.\" I have to establish the local host connection manually.

Beyond that Earthlink has been singularly unforthcoming about exactly what they did...except that what they pushed down the pipe wasn\'t itself stable until January 13 at which time, I uninstalled the entire Earthlink package and reinstalled it.

I\'m starting to get the feeling that I need to do the same with Boinc; that something in CPDN (or Boinc itself) got corrupted during my Earthlink adventure and simply detaching and reattaching wasn\'t sufficient to correct it.

Does Boinc and its scientific clients uninstall cleanly, or do they require registry edits to truly scour them off the system?

Thanks

Lynn

---

Sorry...I forgot to add that I run Boinc in a partition that has a bit over 7 gigabytes of free space. In the past (when it was running properly) CPDN occupied about 40 megabytes in that partition (which is not my \"C:\" partition)

I have 1Gig of RAM and my preferences are set to leave CPDN in memory when Boinc swaps between CPDN and Seti. A preference scheme which had been working quite nicely for a couple of months.

Thanks again,

Lynn

---

There\'s not much in the registry, doing a normal uninstall then deleting the directories would clear it off.

Make sure that you have made a record of all your connection details first (the URL, your username, email and password that you used for this account).

---

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

... where do I find the \"boinc Manager\'s Messages tab\"

Lynn

It\'s one of the \"tabs\" or headings in the Manager.
They\'re called tabs, because they\'re meant to resemble the pieces of cardboard or plastic that stick up from the tops of office folders that are stored in filing cabinets.

The Manager is the rectangular window that opens when you click on the round icon with the B in the bottom right hand side of the display screen.

The names along the tabs are, from the left: Projects, Work (or Tasks) Transfer, Messages, Statistics, Disk.

After two months of adjustments, Earthlink\'s security console seems stable and I\'ve reloaded my system, including the most current version of the Boinc software for Windows. Many of the Boinc problems I\'ve been experiencing since January are now resolved, but I\'m still unable to run the climate-prediction software.

Boinc downloads a new unit, runs it for a few minutes, and then abandons the effort with a raft of error messages which are pretty much identical to the ones I was collecting before. (I\'ve collected an example in a text file which I could send, if that would help).

Earthlink\'s firewall is set to accept data in-and-out from Boinc Manager for Windows, Boinc screensaver, and Boinc client utilizing any port and either the TCP or the UDP protocol. (I\'ve noted no climate-prediction-specific data requests) Is there some way I can determine if climate prediction downloads are being compromised or corrupted as they pass through the Earthlink firewall.

At this exact moment, the last two messages in my message queue are:
3/14/2007 1:04:03 PM|climateprediction.net|Starting hadcm3ohe_0tjc_05686618_0
3/14/2007 1:04:03 PM|climateprediction.net|Starting task hadcm3ohe_0tjc_05686618_0 using hadcm3 version 515

and the software has been running for approximately 25 minutes. In the Tasks panel of my Boinc manager software, I can see that climate prediction is running, but the CPU time column remains zeroed out.

Is there anything here that hints at a solution for my problems?

Thanks!

Lynn Abbey

---

Looking at the last two crashes, both after about a minute, I see no \"smoking gun.\"
What version of boinc are you using?
Could be an unstable machine. Is it overclocked? Running Prime95 Torture Test for ~24 hours should demonstrate stability -- or weaknesses.
http://www.mersenne.org/freesoft.htm

[Edited for typo.]

---

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

I\'m using BOINC client version 5.8.15 for windows_intelx86 which I got from Boinc over the weekend.

I\'m not overclocking, but I will run the torture test as you suggest...

Thanks,

Lynn

Looking at the last two crashes, both after about a minute, I see no \"smoking gun.\"
What version of boinc are you using?
Could be an unstable machine. Is it overclocked? Running Prime95 Torture Test for ~24 hours should demonstrate stability -- or weaknesses.
http://www.mersenne.org/freesoft.htm

[Edited for typo.]

---