1) Forums : Technical Support : Vbox not working here (Message 21908)
Posted 16 hours ago by mmonnin
Post:
Just for S&Gs, can you manually install an OS with VBox. Just to separate physical capability from BOINC detection.

Maybe an older version of Vbox. v5.1 something. Or an older version of BOINC.



5.1 is not supported. 5.2.8 is what is supplied by BOINC. Installed 5.2.10 and related extension pack to see if any change will happen. Right now my cache is full of work, so maybe when BOINC tells Cosmo to go look for more work I will see a change. I am highly skeptical of that.


Yes it is. I am running the vbox app here with 5.1.22
http://www.cosmologyathome.org/show_host_detail.php?hostid=309417

If you want help follow the suggestions.
2) Forums : Technical Support : Vbox not working here (Message 21903)
Posted 3 days ago by mmonnin
Post:
Just for S&Gs, can you manually install an OS with VBox. Just to separate physical capability from BOINC detection.

Maybe an older version of Vbox. v5.1 something. Or an older version of BOINC.
3) Forums : Technical Support : No tasks (Message 21887)
Posted 4 days ago by mmonnin
Post:
Docker is out again.
4) Forums : Technical Support : No tasks (Message 21880)
Posted 6 days ago by mmonnin
Post:
I don't see any docker tasks still.
5) Forums : Technical Support : No tasks (Message 21872)
Posted 8 days ago by mmonnin
Post:
It's running now as is everything listed. Purge isn't something that probably needs to run all the time.
6) Forums : Technical Support : No tasks (Message 21869)
Posted 9 days ago by mmonnin
Post:
No more docker tasks?
7) Forums : Technical Support : Snooze function - options request (Message 21860)
Posted 22 days ago by mmonnin
Post:
A GPU can be used in Linux w/o a desktop environment (GUI).
8) Forums : General Topics : Formula BOINC sprint at Cosmology@Home, July 2018 (Message 21851)
Posted 25 days ago by mmonnin
Post:
Yeah cause F1 runs on equal ground when every car ends up a lap down except for the top 3 teams. And when the leader can lose 160HP and still no one can pass him.

If projects don't want to participate they can ask to not be included. FB is NOT NEW. It's been running for over a decade.


Maybe but how can a Project be excluded if they are never asked in the First Place?
Also by "equal ground" being they all get 100 Gallons of Full not some getting 100 Gallons and some only 50 Gallons of Full.


Because its over a decade old. That's how to not be included. FB is not new.

People complain every week about it. It's getting old.
9) Forums : Technical Support : VM Hypervisor failed to enter an online state in a timely fasion (6 CPUs) (Message 21837)
Posted 26 days ago by mmonnin
Post:
Since you have no completed camb_boinc2docker at all, it is pretty clear that virtualization is not enabled for your PC. So VirtualBox times out, and that is the error message you get.

Boot into your motherboard BIOS, and see if you can find Vt-x and enable it. That should fix it, but if not, ask back and their are further steps you can take.



Host details says this:

Virtualization Virtualbox (5.2.8) installed, CPU has hardware virtualization support and it is enabled

Either out of memory to run the tasks or picky VB version.
10) Forums : General Topics : Formula BOINC sprint at Cosmology@Home, July 2018 (Message 21794)
Posted 29 days ago by mmonnin
Post:
Yeah cause F1 runs on equal ground when every car ends up a lap down except for the top 3 teams. And when the leader can lose 160HP and still no one can pass him.

If projects don't want to participate they can ask to not be included. FB is NOT NEW. It's been running for over a decade.
11) Forums : General Topics : Formula BOINC sprint at Cosmology@Home, July 2018 (Message 21789)
Posted 29 days ago by mmonnin
Post:
I started up this project again the other day and it seems like only 500 of each app were in the queue. Now its dry.

The server is going to have a bad day.
12) Forums : Technical Support : C@H hit or miss... (Message 21731)
Posted 8 Jun 2018 by mmonnin
Post:
camb_boinc2docker will use 2Gb per running task. It doesn't matter if you have it assigned one or more processors. It may cap out at 8 cores per task and I don't have a way to test larger. My tasks run about 5 to 5.5 minutes using eight cores on an AMD Ryzen 7 1700 model.

You may be better off assigning a project to each computer instead of mixing them.


We are mixing projects as Vbox apps waste less resources with fewer cores setting up the app but then they use too much memory at 1-2 cores per task. 23min run time, 40min CPU time on 2 threads. About 5-6min CPU time wasted.
13) Forums : Technical Support : C@H hit or miss... (Message 21729)
Posted 7 Jun 2018 by mmonnin
Post:
Thats the same way it works at LHC with this option. Jobs is the # of tasks that can be downloaded (from waiting to run to waiting to report) at once.

I think CPUs should limit the # of cores.

To run concurrently with another project, you may have to lower the C@H resource share so BOINC will get some WCG tasks. I've ran in issues trying to limit max concurrent tasks where I end up with idle threads as BOINC didn't download any other work. Once the share becomes more equal it can be raised to match WCG.

I am currently running TN-Grid and C@H (with limited threads for VBoX) and it eventually leveled out to run equal amounts of each project.
14) Forums : Technical Support : Legacy tasks not advancing (Message 21723)
Posted 6 Jun 2018 by mmonnin
Post:
I have gotten some as well, just on one machine though. Some ran for several days. I thought they were just long tasks but after pausing, restarting they ended up going back to a lower percentage. I ended up aborting them.

http://www.cosmologyathome.org/workunit.php?wuid=54811306
http://www.cosmologyathome.org/workunit.php?wuid=54776206
http://www.cosmologyathome.org/workunit.php?wuid=54765590
http://www.cosmologyathome.org/workunit.php?wuid=54726945
http://www.cosmologyathome.org/workunit.php?wuid=54682216
http://www.cosmologyathome.org/workunit.php?wuid=54682159
15) Forums : News : Website tweaks (Message 21716)
Posted 4 Jun 2018 by mmonnin
Post:
A forced refresh with Control + F5 updated it for me.
16) Forums : Technical Support : C@H hit or miss... (Message 21715)
Posted 4 Jun 2018 by mmonnin
Post:
Host IDs can be found here too. To me thats a bit simpler.
https://www.cosmologyathome.org/hosts_user.php?userid=1482829

What does the event log say when doing an update for the project on one of these hosts?
17) Forums : News : Website tweaks (Message 21711)
Posted 3 Jun 2018 by mmonnin
Post:
There is a bug in the display.

The pulldown lists for "Community" , "Server" and "Stat" are permanently shown over what ever other screen is selected.

Example, I cannot change resourceshare in projectpreferences because the datafield is behind de pulldownlist.


I have that on one computer but it works on another. Both in Chrome. So I assumed it on that PC. Are the icrons in front of the thread titles indicating the thread has been read or not really tiny as well?
18) Forums : Technical Support : VM Hypervisor (Message 21704)
Posted 31 May 2018 by mmonnin
Post:
With only 8GB of RAM you probably ran out of memory. I think 2 threads per task was using almost 2GB as BOINC sees it. Sometimes the full memory usage is not reported to BOINC.

LHC has mostly VBpx apps. Only SixTrack does not use VBox at LHC. I can run VBox apps from both projects (and more) on several computes.
19) Forums : News : Website tweaks (Message 21700)
Posted 30 May 2018 by mmonnin
Post:
The changes had introduced a typo in the upload URL which should be fixed now. Are you guys seeing the jobs upload correctly now the next time your client attempts to do so?


Yes, they have uploaded now. Thanks.
20) Forums : News : Website tweaks (Message 21696)
Posted 30 May 2018 by mmonnin
Post:
Can anyone else upload tasks?


Next 20