I recently added World Community Grid to my BOINC projects. This used to run as a separate program, but they joined BOINC. Ever since, I noticed SETI is constantly restarting its projects. Is this due to a setting that needs to be changed?

---

If you are seeing messages like:

7/20/2008 8:18:32 PM|SETI@home|Restarting task 04fe08ab.28725.4980.13.8.90_1 using setiathome_enhanced version 528
7/20/2008 8:19:57 PM|SETI@home|Starting 11mr08ab.19322.28648.14.8.225_0
7/20/2008 8:19:57 PM|SETI@home|Starting task 11mr08ab.19322.28648.14.8.225_0 using setiathome_enhanced version 528

in your Messages tab, this is perfectly normal when doing multiple projects. The "restarting" message actually means that it is resuming where it last check-pointed and should pick up where it last left off. It may sometimes look like it starts over from zero while it recalculates how much time is left, but then it should jump right to where it left off.

The messages above are from my own system that I know is running perfectly fine.

---

No, this is not normal. And yes, it does restart the task from the beginning, never finishing it. Here is a copy of my current message log:

2008-07-20 20:58|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-20 21:24|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-20 21:30|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-20 22:03|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-20 22:09|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-20 23:17|World Community Grid|Restarting task R00053_5ef936015001bacd5ba4d9025d93ced6_01_000_15 using rice version 617
2008-07-21 03:12|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 04:12|World Community Grid|Restarting task R00053_5ef936015001bacd5ba4d9025d93ced6_01_000_15 using rice version 617
2008-07-21 05:16|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 06:16|World Community Grid|Restarting task R00053_5ef936015001bacd5ba4d9025d93ced6_01_000_15 using rice version 617
2008-07-21 07:21|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 07:39|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 07:44|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 08:32|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 08:38|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 09:25|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 09:59|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 10:19|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 10:52|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 11:12|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 11:17|SETI@home|Restarting task 29my08af.960.5798.8.8.68_1 using setiathome_enhanced version 527
2008-07-21 12:17|World Community Grid|Restarting task R00053_5ef936015001bacd5ba4d9025d93ced6_01_000_15 using rice version 617

---

In your preferences, do you have set xx in Use at most xx percent of CPU time to less than 100%? If so, reset it to 100%. The "CPU throttling" is known to cause restarting of tasks on some machines. If you get heat or response time problems when running on full load, set On multiprocessors, use at most xx processors to 1 in your preferences. That may run only one task at a time, but will prevent your computer from overheating and all your tasks from continuous restarting.

A fix for this problem (better CPU throttling) will be added to the 6.4 client.

---

In your preferences, do you have set xx in Use at most xx percent of CPU time to less than 100%? If so, reset it to 100%. The "CPU throttling" is known to cause restarting of tasks on some machines. If you get heat or response time problems when running on full load, set On multiprocessors, use at most xx processors to 1 in your preferences. That may run only one task at a time, but will prevent your computer from overheating and all your tasks from continuous restarting.

A fix for this problem (better CPU throttling) will be added to the 6.4 client.

I was just about to suggest quite the same :-)

Gruß,
Gundolf

---

Computer sind nicht alles im Leben. (Kleiner Scherz)

SETI@home classic workunits 3,758
SETI@home classic CPU time 66,520 hours

Ok, thank you both. I made the change (it was set to 50%.) I suspended the current World Community Grid task so the SETI task would run. I will monitor it and see if that does it. Thanks again and congrats Ageless on 11 months and 2 days! ;-)

---

Well, it still restarted, even after making that change.

---

Well, it still restarted, even after making that change.

Did you update the preferences locally, or through the web site?
If through the web site, did you also update Seti after you saved the changes to the web site?

---

Through the website, and yes, I updated after. But now it's been several hours and still doing the same thing.

---

Could you look in both global_prefs.xml and global_prefs_override.xml to see what this value is:

<cpu_usage_limit>100</cpu_usage_limit>

We might as well go directly to the data used to see what the effective setting really is.

---

BOINC WIKI

Ok,the value in global_prefs.xml was 100, but the value in global_prefs_override.xml was 50. I changed it to 100 also and resaved it through notepad. I then updated the project through BOINC. I'll let you know how it goes. Thanks.

---

Ok,the value in global_prefs.xml was 100, but the value in global_prefs_override.xml was 50. I changed it to 100 also and resaved it through notepad. I then updated the project through BOINC. I'll let you know how it goes. Thanks.

Stop and start BOINC or in more recent versions Advanced -> Read Local Prefs File. The override file is for local settings and supersedes whatever is in the global_prefs.xml file.

---

BOINC WIKI

Ok, I went to Advanced -> Read Local Prefs File. So far the task has made it to 67% completion rate, the highest I have seen since this problem started. So I believe the problem is corrected, but I will leave this thread open until at least the first task is completed. Thanks.

---

Ok, it seems to be working properly now. Thanks everyone!

---