[

@Petri
It sounds like the bug is carrying through in all newer versions. Don't know what your time constraints are, but myself and many others who have the 700 series cards, would certainly appreciate it if you could hunt down the problem with the app so that we could move up to the newer Cuda 8.0 versions. Your cuda optimizations are great, but it is going to kinda suck if everybody can't use them.
I'm new to Linux, but if I can help with testing, let me know.

Thanks Guys.

I'm sorry. I have only 1080 and 1080Ti cards. So I can not test with 780.
I have time on holidays, weekends and summers. I could take a look at the code and see if there is something obvious like shared mem size, specific instructions to 980 or 1080 series, etc next time I have time.

Petri

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

Doesn't work for me that way at all. The MAXIMUM tasks I have ever downloaded at one time is the grand total of 22. More normal are downloaded bunches of 14 or 16. It takes me easily over an hour to replenish the cpu cache after rescheduling.

I have come to the conclusion that it has something to do with your resource shares between the other projects that you always seem to have on your computers. I know we have gone though all your cc settings and what we could think of before with no changes for you. When I reschedule CPU ->> GPU 99.8% of the time I get the full cache restored in one shot, which is 90-94 tasks, depending on the computer. So try removing your other projects ... or set them to 0 priority, and no new tasks - they don't seem to affect me that way.

Yes, I suspect my problem probably has something to do with running other projects on the Windows machines. Still does not explain why I have the problem on the new linux cruncher which has never run anything other than SETI. Also, I have not run any other projects other than SETI from 2 days before the start of the WOW contest. I have set NNT on my other projects and have no tasks of any other kind on board the Windows machines other than SETI.

I have not seen any change in number of tasks downloaded after changing the max download per project to 8 as I suspected would have no effect. I ran out of GPU tasks this morning apparently on the linux cruncher since I have a couple of CPU tasks waiting to run and that would only happen if no GPU tasks were available and the system ran CPU tasks on all cores. I am slowly recovering GPU work on that machine in my normal 14 task download grabs. So, after running out and requesting work, I certainly didn't get the 300 tasks for my cache on the first request for work with no work onboard.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I think 10GB is big enough for BOINC to download into. I only ever ran into that problem when I let Einstein download willy-nilly. Certainly has never been the case for SETI.

. . Is that what BOINC is showing when you select the Disk tab? Because that should definitely be more than enough.

Stephen

?

Yes, or about 500 MB less than the allotted 10GB.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

. . Hi Laurent,

. . If that is CPU2GPU then I am using that and it works well enough, but it has two limitations that I find irksome. It only moves tasks in one direction as the name says and it is not selective. It moves all or none. But it does allow me to stow some extra tasks to try and get through the outage without work starvation.

:)

No problem to add more features ! Moving GPU2CPU is a good idea to download more wu in one shot.
Do you want an option to move only some tasks ? based on wu name ? number of tasks ?

---

. . Hi Laurent,

. . If that is CPU2GPU then I am using that and it works well enough, but it has two limitations that I find irksome. It only moves tasks in one direction as the name says and it is not selective. It moves all or none. But it does allow me to stow some extra tasks to try and get through the outage without work starvation.

:)

No problem to add more features ! Moving GPU2CPU is a good idea to download more wu in one shot.
Do you want an option to move only some tasks ? based on wu name ? number of tasks ?

. . Hi again,

. . Moving in both directions is the biggy, but with Stubbles' script you can select the tasks to be moved by suspending them first. That would be great for me as I am used to that method. But I cannot speak for others ...

Stephen
:)

I know that many others receive hundreds or at least tens of task per download. I have never and have always been envious of others that seem to have the golden finger. I'm positive I have tried the cc_config setting you mention. But I once again have set it to 8 and will see it if has any effect. What's the definition of insanity .... doing the same thing exactly the same way and expecting a different result.

Hi Keith,

This is what I have in my cc_config:

<max_file_xfers>20</max_file_xfers>
<max_file_xfers_per_project>20</max_file_xfers_per_project>

placed in the Options section.

With this I download twenty at a time (max number of resends).
I can receive a hundred or more work units at a time, don't know why you get so few. Maybe your ISP is limiting you?
Try this, it's worth a shot.

---

Bruce

...Still does not explain why I have the problem on the new linux cruncher which has never run anything other than SETI.

Hey Keith, I've been thinking (dangerous I know),
Remember way back when you were having issues with Einstein downloading many, many tasks ... I ran across an option/feature in BoincTasks to limit downloads and mentioned it to you. (You discovered it was global, not project specific). Did you ever try implementing that feature? It could be possible that it is BoincTasks causing your limit problems.

Wow, that does sound vaguely familiar. I need to poke around BoincTasks now I guess. That would be the only thing all machines have in common other than internet connection and running SETI.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I found my PM to you about it .. <snip>
WU regulator
As of 1.43, this function can now be found in: C:\Program Files\eFMer\BoincTasks\examples config.xml.
Instead of having a work buffer for a number of days, BoincTasks has the option for a work buffer of xx Work Units.
EDIT: Mine is set to 5000

I have that config.xml file in the /examples directory. But the file says you have to move it next to the BoincTasks executable to make it active. Mine has the max tasks set to 5000 also. I do not have that file in the directory containing the BoincTasks executable.

However I seem to have config.xml.old and config.editor.xml in the executable directory. So that would infer there was at some time the config.xml file controlling the BoincTasks executable.

This is config.xml.old

<config>
<refresh>
<auto>4000</auto>
</refresh>
<options>
<no_cleanup>1</no_cleanup>
</options>
<regulator>
<max_tasks>400</max_tasks>
<days_work_requested>5</days_work_requested>
<cpu_cores>12</cpu_cores>
<gpu_cores>4</gpu_cores>
<min_days_gpu>0.5</min_days_gpu>
<min_days_cpu>0.5</min_days_cpu>
<days_work_requested_min>4.0</days_work_requested_min>
</regulator>
</config>

That would seem to have placed a max of 400 tasks in place. That is a lot less than the 5000 in the examples directory file. I wonder if the .old file has the same effect that the config.xml file is supposed to have.

The other file seems to be a long list of all the logging options possible for cc_config.xml. I did remember using BoincTasks back in May to edit my cc_config down to minimum. That is when I found that my cc_config got returned to max size and options. I only learned this week that later BOINC versions fully populate the cc_config and any editing of reduction in options is for naught since it will be re-written to standard at every BOINC relaunch.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

. . Also, are you running Linux and the repository version of BOINC? Because there is a script you can run that restarts the BOINC client as a service without rebooting the system. It has saved me a lot of irritation on my rig that is running the above configuration. It is a small simple script.

. . You could even just issue the commands form the terminal each time.

:(

Yes I'm running repository BOINC, but I'm using Mint and I think I've read you can't stop the service without rebooting. Otherwise a script seems perfect.

. . Also, are you running Linux and the repository version of BOINC? Because there is a script you can run that restarts the BOINC client as a service without rebooting the system. It has saved me a lot of irritation on my rig that is running the above configuration. It is a small simple script.

. . You could even just issue the commands form the terminal each time.

:(

Yes I'm running repository BOINC, but I'm using Mint and I think I've read you can't stop the service without rebooting. Otherwise a script seems perfect.

'service boinc stop' to stop
'service boinc start' to start
'service boinc status' to get the status

---

I use the repository version as well on Mint and Ubuntu and restarting BOINC is easy
/etc/init.d/boinc-client restart

or
/etc/init.d/boinc-client stop
/etc/init.d/boinc-client start
EDIT: You need root access, so 'sudo' them.

Not sure there Keith if BoincTasks could be reading the .OLD file or not.

I guess one way to test would be to hard disconnect BonicTassks from the network (pull the ethernet, or shutdown your daily computer) then reboot and try a reschedule in Linux and see if it acts differently.

OK, Gents .... here's an update. Since I didn't know if that .old file had any effect, I just deleted it and the other one in the BoincTasks directory. Stopped and started BoincTasks on my daily driver which monitors all my crunchers from a central location. Didn't seem to have any effect on BoincTasks.

Noticed about 5 minutes ago that Darksider, my special app box was down about 150 GPU tasks and had been getting no work available responses for the past hour. So I did the NNT and Suspend Network Activity trick and waited out the 5 minute connect interval. And look at this:

Sun 27 Aug 2017 04:21:19 PM PDT | SETI@home | work fetch resumed by user
Sun 27 Aug 2017 04:21:39 PM PDT |  | Resuming network activity
Sun 27 Aug 2017 04:21:49 PM PDT | SETI@home | Sending scheduler request: To fetch work.
Sun 27 Aug 2017 04:21:49 PM PDT | SETI@home | Reporting 15 completed tasks
Sun 27 Aug 2017 04:21:49 PM PDT | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
Sun 27 Aug 2017 04:21:54 PM PDT | SETI@home | Scheduler request completed: got 146 new tasks

Sun 27 Aug 2017 04:27:00 PM PDT | SETI@home | Reporting 8 completed tasks
Sun 27 Aug 2017 04:27:00 PM PDT | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
Sun 27 Aug 2017 04:27:03 PM PDT | SETI@home | Scheduler request completed: got 56 new tasks

So I think I may have FINALLY figured out why I never got more than 22 tasks at one time per download. That .old file with the 400 task max setting must have been hamstringing me all along.

A BIG thanks to Brent for digging into his PM history and finding that message which has given me the information to fix my download issues.

Thanks to all for all your help and specially your patience with my whining and whinging.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Is that a "BIG thanks" for remembering our discussion about it 3 months ago (May 19th by PM) or an ...
<eye_roll><pissed_off_look>BIG thanks</pissed_off_look></eye_roll> for mentioning it in the first place and causing many hours of frustration?

Anyways,
Cheers, Glad it's working now!

LOL, Brent. Naw that is just a big 'ole hug of thanks. We just saw it work on the linux box but the jury is still out on my Windows7 daily driver. I am down about 35 tasks now with nothing but you have reached your limit messages. So, NNT and stop network activity. Wait it out and restart and set tasks and network back in place. Make work request and ....... get ONE task.

I hope that is just that the buffer had nothing but VLARs in it an the next request will fill out my cache with 35 or so tasks. That will prove the problem and fix. Wait and see ......

[Edit] Next request only got 10 tasks. Probably won't prove anything till Monday night after rescheduling.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Whew....... I think I have my issue licked. This is from my daily driver which was still getting spurts of 14 tasks downloaded. I rebooted the machine since my last post.

8/27/2017 18:34:46 | SETI@home | Sending scheduler request: To fetch work.
8/27/2017 18:34:46 | SETI@home | Reporting 2 completed tasks
8/27/2017 18:34:46 | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
8/27/2017 18:34:49 | SETI@home | Scheduler request completed: got 33 new tasks

Most ever on that machine so far.... or since things went sideways back in May {grin} Looking evilly at you Brent.. Ha Ha!

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Stand back I'm thinking again,
Now that we have proven how tightly integrated BoincTasks and BOINC can become when it comes to scheduling requests, it brings to mind that some people have problems getting tasks and some don't - Is it BoincTasks that is causing it????

Does the scheduler request go through BT before it is sent to the server, or are the setting loaded when network communication is established?
What if .....

Has me thinking too. There has to be some 'hooks' into the Manager to effect BoincTasks task control obviously. So where does BoincTasks intercept the Manager-Scheduler server communication? Is there some 'global' hidden setting that it sets that is not exposed in the Manager interface?

Looks like we need to post some questions to Fred over in the BoincTasks forums.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)