Ever since July 16, I am getting "Error while computing" for each WU downloaded. I have tried to "reset project" and it downloads another, but same result.

I'm running BOINC v 6.12.33 PowerPC, wxWidgets v 2.8.10

I'm wasting work units... Help?

Thanks,
Margo

All your tasks end with: error -177 "Maximum elapsed time exceeded" after less than 20 seconds CPU/elapsed time.

That might be caused by that computer's Average processing rate of 122107.58946984, which looks a bit high to me.

Since that is a server-side setting, I can't tell you how to change that, if you get no chance to successfully complete a single task.

What is the Duration Correction factor for SETI@home on that computer? (Projects tab->right click->properties)

Gruß,
Gundolf

Hi,

The Duration correction factor on the server is 1.0000

Scheduling is set as:
CPU scheduling priority 0.00
CPU work fetch priority 0.00
CPU work fetch deferred for 00:08:48
CPU work fetch deferral interval 00:10:00

Is there anything else I can check to help?

THANKS,
Margo

You seem to have no tasks downloaded. Do you have "No New Tasks" set?

Without successfully completed tasks, your Average Processing Rate will never adjust.

Are you comfortable editing the client_state.xml file? That's no easy task, but I know of no other way to get tasks finished on your computer, since I'm not sure if resetting the project or detaching/reattaching would help with the APR.

Perhaps someone else is more experienced than I am.

Gruß,
Gundolf

I'm not sure if resetting the project or detaching/reattaching would help with the APR.

Perhaps someone else is more experienced than I am.

Not pretending to be "more experienced" than you ;)
But maybe the Detach/re-Attach will give new "Computer ID" to start fresh?


By editing the client_state.xml (using simple text editor)
maybe you mean finding and changing the duration_correction_factor

...
<project>
<master_url>http://setiathome.berkeley.edu/</master_url>
<project_name>SETI@home</project_name>
...........
<duration_correction_factor>1.000000</duration_correction_factor>


Changing the value from 1.000000 to e.g. 10000.00 may make BOINC adjust the estimated times (from 1 estimated second as it is now to 10000 estimated seconds)?

---

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 

Hi Gundolf & BilBg,

First, thank you so much for your suggestions and help. I am comfortable with editing the client_state.xml, but it does not exist on my Mac. Is it hidden? A search comes up empty.

"No new tasks" is not set so I can get new tasks.

I reset the project, but that did not help.

I removed the project and then added it, but that did not resolve the problem.

So very odd.

Thanks,
Margo

I am comfortable with editing the client_state.xml, but it does not exist on my Mac. Is it hidden? A search comes up empty.

That's strange. On a windows computer, I'd say it's hidden, but I don't know if that's possible on a Mac. (It's been decades...;-)

"No new tasks" is not set so I can get new tasks.

But you are not getting any. The last one reported is still from 13 Aug 2011 | 11:13:56 UTC. Does the button say "No new tasks" or "Allow new tasks"?

I reset the project, but that did not help.

I removed the project and then added it, but that did not resolve the problem.

So very odd.

Indeed, but that's what I meant with 'I'm not sure if resetting the project or detaching/reattaching would help with the APR'. :-)

If you happen to find your client_state.xml file, you could force your computer to get a new hostID. That way, the possibility of resetting the APR would be higher. By setting <rpc_seqno> in the seti@home section to a lower value, you'd force a new hostID (IIRC). But make sure to change only the seti@home section, otherwise your CPDN ID would be reset also, or worse.

Gruß,
Gundolf

The location of the file on the Mac is HD/Library/Application Support/BOINC Data/Client_State.xml .

---

Thank you for the location of the XML file. I edited it to change the duration_correction_factor and shut down and restarted BOINC but there was no change.

The No New Tasks button is labeled "No new tasks" and the status indicates that this means I will allow tasks.

I added 2 projects; climateprediction and Enigma and learned something interesting that may be key: climateprediction warned me that the project may not run on my computer (I added it anyway) and Enigma had no such warning. The result? Enigma is running and processing tasks and climateprediction is deferring communication just like seti@home.

The Mac I am running this on is a Dual 1.8 GHz PowerPC G5. Old technology. Perhaps that is the answer?

Thank you all for your support in this endeavor; this user community is the best!
Margo

Or maybe it has something to do with this?

===== message on home page =====================================================
work generation temporarily halted
The RAID that holds our raw data is down, so the splitters have nothing to work on. New work generation will be at a trickle until this is fixed. 19 Aug 2011 | 16:59:10 UTC · Comment
================================================================================

??
Margo

I edited it to change the duration_correction_factor and shut down and restarted BOINC but there was no change.

The correct sequence is: shut down BOINC, edit client_state.xml, restart BOINC. If BOINC is running while you change the file, it will revert your changes immediately.

The No New Tasks button is labeled "No new tasks" and the status indicates that this means I will allow tasks.

The reason why your computer isn't getting new tasks is probably that none are available due to the disk RAID problems, but my little laptops have downloaded 5 tasks nevertheless, so I just wanted to be sure.

The Mac I am running this on is a Dual 1.8 GHz PowerPC G5. Old technology. Perhaps that is the answer?

I don't think so. One of my laptops is a 700 MHz Pentium III and it works just fine with SETI (not with CPDN though). You'll just have to wait until new work is generated.

Gruß,
Gundolf
[edit]Yes, that's the main reason. You just posted your assumption while I was composing this message:-)[/edit]

Thanks for your help, Grundolf.

Margo

Longer explanation of the same by Richard Haselgrove (very reputable "guru" here ;) ):

(How to get new HostIDs to normalize the APR by editing <rpc_seqno>nnnnn</rpc_seqno> in client_state.xml file)
http://setiathome.berkeley.edu/forum_thread.php?id=65291&nowrap=true#1145786

(as usual - know that the first click on the above link will not lead you to the right place,
use the link second time to see the Richard Haselgrove's post #1145786)

---

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)