Do we have limits again?

19-Jul-10 18:04:13	SETI@home	Requesting new tasks
19-Jul-10 18:04:39	SETI@home	Scheduler request completed: got 0 new tasks
19-Jul-10 18:04:39	SETI@home	Message from server: No work sent
19-Jul-10 18:04:39	SETI@home	Message from server: This computer has reached a limit on tasks in progress

---

Who the hell is General Failure and why is he reading my harddisk?¿

Yes - see tech news.

I recognize that we are not yet managing these server runs with clear, measurable, goals to aim towards. And that it would be good to have such goals. It's been a bit seat of the pants thus far.

---

Yes - see tech news.

I recognize that we are not yet managing these server runs with clear, measurable, goals to aim towards. And that it would be good to have such goals. It's been a bit seat of the pants thus far.

Just making your link clickable, Jeff!

Yes - see tech news.

I recognize that we are not yet managing these server runs with clear, measurable, goals to aim towards. And that it would be good to have such goals. It's been a bit seat of the pants thus far.

Just making your link clickable, Jeff!

I finished my Downloads, servers should be fine now...LOL

---

Official Abuser of Boinc Buttons...
And no good credit hound!

my computer is back to thinking it has 10+ hour work units.. actual will be around 3 hour..

*sigh* I still think I have enough. Oh well. I am definately out of the rush for more either way.

---

Janice

my computer is back to thinking it has 10+ hour work units.. actual will be around 3 hour..

*sigh* I still think I have enough. Oh well. I am definately out of the rush for more either way.

After they lift the quota again, just reset you DCF to .2 and it should ask for work. I had to set mine not to ask for tasks as I have plenty now and it wanted more...

---

Official Abuser of Boinc Buttons...
And no good credit hound!

my computer is back to thinking it has 10+ hour work units.. actual will be around 3 hour..

*sigh* I still think I have enough. Oh well. I am definately out of the rush for more either way.

After they lift the quota again, just reset you DCF to .2 and it should ask for work. I had to set mine not to ask for tasks as I have plenty now and it wanted more...

With soft^spirit running stock, setting DCF to .2 might cause downloading about 15 days of work for a 3 day cache setting. Dividing the project DCF by about 3.5 would be a reasonable correction, though. That would get the 10+ estimates back to around 3 and encourage about the right amount of work fetch.

Let's also try to keep clear the difference between the daily quota (Max tasks per day scaled) and the limits on how many "in progress" tasks there can be before the Scheduler refuses work. In times like these there are likely to be many users checking the forums who normally don't. Not necessarily newbies, but quite likely unfamiliar with the recent changes in BOINC and how the project is controlling work flow.

                                                               Joe

Yes - see tech news.

I recognize that we are not yet managing these server runs with clear, measurable, goals to aim towards. And that it would be good to have such goals. It's been a bit seat of the pants thus far.

Why not to decrease the max. WU cache size from 10 to 5 days?

I guess then we wouldn't see so much (maxed out) traffic and all hosts have the change to fill up a 5 day WU cache for to bridge planned and unplanned outages.

I think it's worth to try it.

---

Yes - see tech news.

I recognize that we are not yet managing these server runs with clear, measurable, goals to aim towards. And that it would be good to have such goals. It's been a bit seat of the pants thus far.

Why not to decrease the max. WU cache size from 10 to 5 days?

I guess then we wouldn't see so much (maxed out) traffic and all hosts have the change to fill up a 5 day WU cache for to bridge planned and unplanned outages.

I think it's worth to try it.

I already ask that question and the response was CUDA card have a problem calculating how much work they have. Setting the limits to 5 days could result in far less or far more than requested. This could be fixed at a latter date but it's not working now.

---

Yes - see tech news.

I recognize that we are not yet managing these server runs with clear, measurable, goals to aim towards. And that it would be good to have such goals. It's been a bit seat of the pants thus far.

Why not to decrease the max. WU cache size from 10 to 5 days?

I guess then we wouldn't see so much (maxed out) traffic and all hosts have the change to fill up a 5 day WU cache for to bridge planned and unplanned outages.

I think it's worth to try it.

I wish the cache setting was per project instead of a BOINC global setting. I'd also like to be able to dedicate resources to a project. In an 8 core system always give Project A at least 4 cores, Project B at least 2 cores, & 2 cores to be used where needed. However, That isn't how they designed the system to work.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

I already ask that question and the response was CUDA card have a problem calculating how much work they have. Setting the limits to 5 days could result in far less or far more than requested. This could be fixed at a latter date but it's not working now.

Not on my system/s. I had/have still flops entries in my app_info.xml file, so BOINC can calculate very well how big the DLed WU cache is.

Maybe stock BOINC/projects (SETI@home) can't calculate well?

---

I already ask that question and the response was CUDA card have a problem calculating how much work they have. Setting the limits to 5 days could result in far less or far more than requested. This could be fixed at a latter date but it's not working now.

Not on my system/s. I had/have still flops entries in my app_info.xml file, so BOINC can calculate very swell how big the DLed WU cache is.

Maybe stock BOINC/projects (SETI@home) can't calculate well?

I wouldn't know because I am unable to run CUDA. I was just repeating what I was told by another user.

---

I already ask that question and the response was CUDA card have a problem calculating how much work they have. Setting the limits to 5 days could result in far less or far more than requested. This could be fixed at a latter date but it's not working now.

Not on my system/s. I had/have still flops entries in my app_info.xml file, so BOINC can calculate very well how big the DLed WU cache is.

Maybe stock BOINC/projects (SETI@home) can't calculate well?

Confirmed. CPU even further off than GPU.

---

Janice

I wouldn't know because I am unable to run CUDA. I was just repeating what I was told by another user.

I guess if the server-FLOPS will be enabled again, then it's possible to reduce the max. WU cache size to 5 days.

---

I've got plenty of work...80+ WU's

---

Confirmed. CPU even further off than GPU.

[offtopic]
Just curious..

Why you don't have optimized project applications installed?
And an app_info.xml file with flops entries?
Double RAC of your CPU and you would protect your GPU for VLAR WUs.

ReSchedule tool recommended because of new WU quota.

If the VLAR WUs will be sent only to the CPU here at SETI@home (what is current tested at SETI@home Beta Test), ReSchedule tool not longer needed.
[/offtopic]

---

To be honest.. I have enough problems with the standard applications that I do not want to compound it further. back "in the day" I ran the standard command line version, and that worked okay.

Just not a big fan of "add on" programs.

---

Janice

is my pc ok for S.E.T.I

it intel Q8300/8gb ram/64 bit windowd vista

why limit still in place? download bandwidth only~50% used...

why limit still in place? download bandwidth only~50% used...

http://setiathome.berkeley.edu/forum_thread.php?id=60810&nowrap=true#1017616