Hi Guys,

I have a problem. I am not getting new workunits.
I read the forums and i checked those thing was written there, but none of those worked for me.

So.... I have Seti and Climateprediction.

At first I've got only SETI but it was down too many times, so I said lets get enough work for 4 days. (change prefs) It was not enough, so I started Climate also to keep my computer busy. It is also was not enough, so I changed my prefs that way to run 80% of the time Climate project and only 20% SETI.

When I saw that I've go ne WU's I suspended climate and was concentrating to Seti.

cca 2 month ago I realized that I am not getting new work for 14 days, than I got work enough for cca 10 days, than another 14 days coming without work.

In my term "a day" means cca 6 hours, thats my standard daily running time of my computer.

Now I see that Seti works fine, so I changed my prefs to share resources between SETI and climate equally. Nothing happens. Now I am on my 14th day and nothing happened. When I checked my points I saw that my average is the same for 14 days without any work????

what I did till today:
- I am pushing the update button frequently
- The trick "no new work" - update - "Allow new work" - update trick also doesn't worked.
- I changed my prefs, nothing happened.
- I also replaced the Seti calculator program to an optimized one. (I know: Theoretically it has no effect to Seti, but what if?? --> nothing happened :) )

These are my latest messages since my computer is on:

2005. 10. 27. 19:16:59||Starting BOINC client version 4.43 for windows_intelx86
2005. 10. 27. 19:16:59||Data directory: J:\\Seti
2005. 10. 27. 19:17:00|SETI@home|Found app_info.xml; using anonymous platform
2005. 10. 27. 19:17:00|climateprediction.net|Computer ID: 205427; location: home; project prefs: default
2005. 10. 27. 19:17:00|SETI@home|Computer ID: 1012183; location: home; project prefs: default
2005. 10. 27. 19:17:00||General prefs: from climateprediction.net (last modified 2005-10-20 20:43:23)
2005. 10. 27. 19:17:00||General prefs: no separate prefs for home; using your defaults
2005. 10. 27. 19:17:00||Remote control not allowed; using loopback address
2005. 10. 27. 19:45:17|climateprediction.net|Deferring computation for result 24ja_000120199_0
2005. 10. 27. 19:45:17||Resuming computation and network activity
2005. 10. 27. 19:45:17||schedule_cpus: must schedule
2005. 10. 27. 19:45:17|climateprediction.net|Restarting result 24ja_000120199_0 using hadsm3 version 4.13
2005. 10. 27. 20:55:17||schedule_cpus: time 4200.031250
2005. 10. 27. 21:22:00|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005. 10. 27. 21:22:01|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
2005. 10. 27. 22:05:17||schedule_cpus: time 4200.031250
2005. 10. 27. 23:15:17||schedule_cpus: time 4200.046875
2005. 10. 27. 23:24:16||Suspending computation and network activity - user is active
2005. 10. 27. 23:24:16|climateprediction.net|Pausing result 24ja_000120199_0 (removed from memory)
2005. 10. 27. 23:24:18||request_reschedule_cpus: process exited
2005. 10. 28. 8:08:58||Resuming computation and network activity
2005. 10. 28. 8:08:58||schedule_cpus: must schedule
2005. 10. 28. 8:09:07|climateprediction.net|Restarting result 24ja_000120199_0 using hadsm3 version 4.13
2005. 10. 28. 8:09:35||Suspending computation and network activity - user is active
2005. 10. 28. 8:09:35|climateprediction.net|Pausing result 24ja_000120199_0 (removed from memory)
2005. 10. 28. 8:09:38||request_reschedule_cpus: process exited
2005. 10. 28. 8:10:43||Resuming computation and network activity
2005. 10. 28. 8:10:43||schedule_cpus: must schedule
2005. 10. 28. 8:10:43|climateprediction.net|Restarting result 24ja_000120199_0 using hadsm3 version 4.13
2005. 10. 28. 9:20:43||schedule_cpus: time 4200.078125
2005. 10. 28. 10:30:43||schedule_cpus: time 4200.046875
2005. 10. 28. 11:40:43||schedule_cpus: time 4200.078125
2005. 10. 28. 12:50:43||schedule_cpus: time 4200.031250
2005. 10. 28. 14:00:43||schedule_cpus: time 4200.031250
2005. 10. 28. 15:10:43||schedule_cpus: time 4200.093750
2005. 10. 28. 15:59:29|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005. 10. 28. 15:59:30|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
2005. 10. 28. 16:20:44||schedule_cpus: time 4200.078125
2005. 10. 28. 17:30:44||schedule_cpus: time 4200.000000
2005. 10. 28. 18:40:44||schedule_cpus: time 4200.015625
2005. 10. 28. 19:50:44||schedule_cpus: time 4200.015625
2005. 10. 28. 21:00:44||schedule_cpus: time 4200.062500
2005. 10. 28. 22:10:44||schedule_cpus: time 4200.000000
2005. 10. 28. 22:40:19||Suspending computation and network activity - user request
2005. 10. 28. 22:40:19|climateprediction.net|Pausing result 24ja_000120199_0 (removed from memory)
2005. 10. 28. 22:40:21||request_reschedule_cpus: process exited
2005. 10. 29. 10:52:47||Resuming computation and network activity
2005. 10. 29. 10:52:47||schedule_cpus: must schedule
2005. 10. 29. 10:52:48|climateprediction.net|Restarting result 24ja_000120199_0 using hadsm3 version 4.13
2005. 10. 29. 10:53:26||request_reschedule_cpus: project op
2005. 10. 29. 10:53:26||schedule_cpus: must schedule
2005. 10. 29. 10:53:27|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
2005. 10. 29. 10:53:29|SETI@home|Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
2005. 10. 29. 10:53:29|SETI@home|Deferring communication with project for 10 minutes and 6 seconds
2005. 10. 29. 10:54:12||request_reschedule_cpus: project op
2005. 10. 29. 10:54:12||schedule_cpus: must schedule
2005. 10. 29. 10:54:14|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
2005. 10. 29. 10:54:16|SETI@home|Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded

---

Two more questions:
- How many Seti WU's are in your work tab?
- What is the deadline of the CPDN WU?

BOINC does not download new WUs after the first completed one, but after a while that depends on benchmarks, cache size and resource shares.
If the deadline of the CPDN WU is in near future, your BOINC might be running in panic mode which forces crunching a single project to reach its deadline.

Leave it alone. All the suspending, no new work, and junk is most likely why it is not working the way you expect it to. Set your preferences the way you want them maybe do one update to get these preferences from the web site and leave it alone.

Other general advice:
Upgrade BOINC to the 5.2.x version that is currently recommended. There have been many improvements since the version you are running, in fact if I remember correctly that version had a bad bug in it that caused it to be pulled within a few days of release.

CPDN has a fairly long period between checkpoints so I would advise either leaving applications in memory or doing work while the computer is in use. That way you will loose less progress when you use the computer.

Set your queue to the size you really want and leave it alone. This is especially true in the 5.2.x versions, after a few completed workunits the client will really get that much work. I recommend the time you expect a local outage to last divided by the number of projects your are attached to as a maximum queue size. For example if you expect your connection to be down 4 days and you are attached to 2 projects then you should set your queue to no more than 2 days, I personally use 0.3 for my queue. A 10 day queue will likely cause EDF mode for seti and then no seti work will be downloaded until the other project(s) catch up.

To sum things up micromanaging BOINC normally makes whatever you are trying to acomplish harder, decide what settings you want and let BOINC adjust to them.

---

BOINC WIKI

BOINCing since 2002/12/8

Leave it alone. All the suspending, no new work, and junk is most likely why it is not working the way you expect it to. Set your preferences the way you want them maybe do one update to get these preferences from the web site and leave it alone.

Other general advice:
Upgrade BOINC to the 5.2.x version that is currently recommended. There have been many improvements since the version you are running, in fact if I remember correctly that version had a bad bug in it that caused it to be pulled within a few days of release.

CPDN has a fairly long period between checkpoints so I would advise either leaving applications in memory or doing work while the computer is in use. That way you will loose less progress when you use the computer.

Set your queue to the size you really want and leave it alone. This is especially true in the 5.2.x versions, after a few completed workunits the client will really get that much work. I recommend the time you expect a local outage to last divided by the number of projects your are attached to as a maximum queue size. For example if you expect your connection to be down 4 days and you are attached to 2 projects then you should set your queue to no more than 2 days, I personally use 0.3 for my queue. A 10 day queue will likely cause EDF mode for seti and then no seti work will be downloaded until the other project(s) catch up.

To sum things up micromanaging BOINC normally makes whatever you are trying to acomplish harder, decide what settings you want and let BOINC adjust to them.

Thanks guys! I really appreciate.

- I will try not to pushing buttons if its not necessary :))
- I will try to forget the "micromanagement" thing. Maybe I played too many RTS games.
- I will also try the upgrade you proposed.


To Ripley:
- Believe me my friend I finished all my workunits cca 16 days ago and I did not seen any new since that. I am empty!

- I dont think that Climate pushed BOINC into "crazy mode".
You said that BOINC maybe try to catch up, because I am not in time.
I don't think so.
I started my work in early August and I've got work for cca 1100 hours. No I reached the point where 473 hours work is done and the deadline is July 2006.
I think I am ahead of schedule and not behind.

---

To Ripley:
- Believe me my friend I finished all my workunits cca 16 days ago and I did not seen any new since that. I am empty!

Well, you're welcome, but this information was missing, that's all :)
Just follow Keck's advices ..

Well, you're welcome, but this information was missing, that's all :)
Just follow Keck's advices ..

Would you tell me whats the best way of upgrading BOINC?

- May I try simply install the new version above the old one?
- or first I have to remove the old one and install the new, than copy back all the work files?

One thing is sure: I will make a backup copy of the BOINC folder...

---

Would you tell me whats the best way of upgrading BOINC?

- May I try simply install the new version above the old one?
- or first I have to remove the old one and install the new, than copy back all the work files?

One thing is sure: I will make a backup copy of the BOINC folder...

Uninstalling a old version does NOT remove the WUs and other data files, because they remain ... so: Uninstall the old stuff and install the new one at the same location.

Uninstalling a old version and installing a new one is always the recommended way for every kind of software except otherwise stated.