23/5-30/5 CERN's 60 years of science LHC T4T@home
Forum rules
#1 23/5-30/5 CERN's 60 years of science LHC T4T@home
http://boincstats.com/en/stats/challenge/team/chat/550
Name first week of celebrating CERN's 60 years of science for peace
Status Upcoming
Project LHC Test4Theory@Home
Issued by CERN
Start time 2014-05-23 00:00 UTC
End time 2014-05-30 00:00 UTC
Late entrants allowed? Yes
Number of teams participating » 12
Number of users participating 0
Name first week of celebrating CERN's 60 years of science for peace
Status Upcoming
Project LHC Test4Theory@Home
Issued by CERN
Start time 2014-05-23 00:00 UTC
End time 2014-05-30 00:00 UTC
Late entrants allowed? Yes
Number of teams participating » 12
Number of users participating 0
The best form of help from above is a sniper on the rooftop....
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#2 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Just a couple of days to go. I've got 2 machines on it & will try for a third.
TSBT actually are quite high flyers in this project, thanks almost entirely to the activities of Derv.
It's a bit fiddly to get started but let's have a few more recruits on it please.
TSBT actually are quite high flyers in this project, thanks almost entirely to the activities of Derv.
It's a bit fiddly to get started but let's have a few more recruits on it please.
Willie the Megacruncher
#3 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
In what way is it fiddly Mega?Megacruncher wrote:It's a bit fiddly to get started but let's have a few more recruits on it please.
Just so I know what to expect
#4 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Does this one not run in vbox ? Never tried it either. Might try and get a puter or two on it as long as my systems stay stable
from the main page
Before Attaching to us you must first have an up to date installed version of the VIRTUAL BOX software package. To download Virtual Box if needed see instructions here.
This project uses BOINC. If you're already running BOINC, select Add Project. If not, download BOINC.
When prompted, enter
http://lhcathome2.cern.ch/test4theory/
If you're running a command-line version of BOINC, create an account first.
Wed, 11/24/2010 - 10:17 — Daniel Lombraña... (updated July 2013, BS)
In order to run the collision simulations you will need to install two key applications in your computer:
VirtualBox
BOINC client.
VirtualBox is a free and open source virtualization tool that allows the creation of virtual machine(s) inside your PC where you can run any operating system. For example, you will be able to run CERN's Scientific Linux system in your Windows computer without risking the integrity of your machine.
LHC@home 2.0 uses this technology to run the complex CERN experiment code in your computer, independently of your OS and hardware. Thus, if you want to collaborate with this project, you first have to install the VirtualBox software, (remember it is free!)
Next you will need to install BOINC (or if you already have it installed, set it up correctly for your platform). BOINC is a standard volunteer computing application that allows you to share your PC(s) with research projects that you choose to attach to. Our project is the first BOINC project that includes virtualization technology to run very complex software in "modest" home computers (well, if you have an eight core machine, you can say it is not modest at all!).
To summarize: if you want to help CERN researchers, you need to first install the VirtualBox software, then install the BOINC software, and finally attach your computer to the project.
from the main page
Before Attaching to us you must first have an up to date installed version of the VIRTUAL BOX software package. To download Virtual Box if needed see instructions here.
This project uses BOINC. If you're already running BOINC, select Add Project. If not, download BOINC.
When prompted, enter
http://lhcathome2.cern.ch/test4theory/
If you're running a command-line version of BOINC, create an account first.
Wed, 11/24/2010 - 10:17 — Daniel Lombraña... (updated July 2013, BS)
In order to run the collision simulations you will need to install two key applications in your computer:
VirtualBox
BOINC client.
VirtualBox is a free and open source virtualization tool that allows the creation of virtual machine(s) inside your PC where you can run any operating system. For example, you will be able to run CERN's Scientific Linux system in your Windows computer without risking the integrity of your machine.
LHC@home 2.0 uses this technology to run the complex CERN experiment code in your computer, independently of your OS and hardware. Thus, if you want to collaborate with this project, you first have to install the VirtualBox software, (remember it is free!)
Next you will need to install BOINC (or if you already have it installed, set it up correctly for your platform). BOINC is a standard volunteer computing application that allows you to share your PC(s) with research projects that you choose to attach to. Our project is the first BOINC project that includes virtualization technology to run very complex software in "modest" home computers (well, if you have an eight core machine, you can say it is not modest at all!).
To summarize: if you want to help CERN researchers, you need to first install the VirtualBox software, then install the BOINC software, and finally attach your computer to the project.
The best form of help from above is a sniper on the rooftop....
#5 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Have posted a brief guide to install here http://www.dunadd.co.uk/seti/forum/view ... 661#p20661
if anyone knows anymore tips or bits I've missed please post up
if anyone knows anymore tips or bits I've missed please post up
The best form of help from above is a sniper on the rooftop....
#6 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Thanks Alez. I'll give it a try when I get home.
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#7 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Do you have to run it under a virtual system or is that just for linux? Running Win 8.1, added the project, it downloaded a WU and is crunching away. App is Test4Theory vboxwrapper 260.76 (vbox64). I don't recall installing virtual box. Estimates over 76 hours remaining on my i7-3770k
Eh, 3 hours in and it's 11% complete.
Eh, 3 hours in and it's 11% complete.
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#8 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Maybe you downloaded vbox at the same time as Boinc. Anyway if it's running it must be installed & 11% in 3 hours sounds about right. :)scole250 wrote:Do you have to run it under a virtual system or is that just for linux? Running Win 8.1, added the project, it downloaded a WU and is crunching away. App is Test4Theory vboxwrapper 260.76 (vbox64). I don't recall installing virtual box. Estimates over 76 hours remaining on my i7-3770k
Eh, 3 hours in and it's 11% complete.
Willie the Megacruncher
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#9 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I had indeed installed it after all. Was looking for it in the wrong folder.
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#10 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Will this project only run 1 task at a time?
By the way, easy peasy install of vbox on Win 7-8. Just download the latest install (I just did 4.3.12), shut down BOINC, install vbox, don't bother starting it, start BOINC, make sure you have enough disk space allocated (needs around 8 GB for T4T, I think), add the project.
If T4T won't download a WU, you may not have enough disk space allocated to BOINC. If it downloads a task and it sits in the waiting to start state, vbox is not installed correctly.
Spoke too soon. Have an AMD machine that had an older version of vbox. Not sure if that messed it up. Can't get the virtual host to run. Give up for night. Will try when I get back in office.
By the way, easy peasy install of vbox on Win 7-8. Just download the latest install (I just did 4.3.12), shut down BOINC, install vbox, don't bother starting it, start BOINC, make sure you have enough disk space allocated (needs around 8 GB for T4T, I think), add the project.
If T4T won't download a WU, you may not have enough disk space allocated to BOINC. If it downloads a task and it sits in the waiting to start state, vbox is not installed correctly.
Spoke too soon. Have an AMD machine that had an older version of vbox. Not sure if that messed it up. Can't get the virtual host to run. Give up for night. Will try when I get back in office.
#11 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I only seem to be able to get 1 task per machine as well. Don't know if it is a project or memory limitation ? Will have a troll through the forums later and see if I can find out. I think each task is going to take me approx 24 hrs to complete. Not really competition material this project.
The best form of help from above is a sniper on the rooftop....
#12 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
From what I can glean T4TH normally only runs 1 unit per host and each task runs for 24 hours. Any idea what credit we can expect from this ? There also seems to be a case of using an old version of boinc 6.something plus an old version of the project wrapper gives double credit
It also seems to be possible to run more than one version of T4TH on each computer, but it involves setting up multiple vbox hosts and boinc clients. Some people seem to have success with this approach, others seem to have problems.
The thread is here http://lhcathome2.cern.ch/test4theory/f ... hp?id=1154
It also seems to be possible to run more than one version of T4TH on each computer, but it involves setting up multiple vbox hosts and boinc clients. Some people seem to have success with this approach, others seem to have problems.
The thread is here http://lhcathome2.cern.ch/test4theory/f ... hp?id=1154
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#13 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Looks like perhaps all tasks, regardless of cpu, run for 24 hours. The faster the cpu, the more credit you get. My i7-3770K got 726 credits.
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#14 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I can see why you say that.Alez wrote:Not really competition material this project.
But I disagree.
The fact that we have Derv, world #8 for the project, with us, ensures that we will not be disgraced. His (or her) solo efforts have been enough to keep us in the top 14 by total credit.
I reckon that, with the new recruits, we can at least match Derv's efforts, even at 1 WU per machine. The new effort, on top of the old should keep us in the top half of teams taking part. So go for it everyone!
Willie the Megacruncher
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#15 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I'm in! Let's give it a go! :bounce: Have to get this working on more systems :)
#16 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
You know my attitude, a scrap is a scrap and top#1 is the goal We will see ....
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#17 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I think I have the multiple boinc client setup going well. Just a long time to see results. Here's the process I used. It's the same as the info in the link alez provided, but it lists it one shot so you don't have to follow the changes through the thread:
For Win7 and Win8...
-Be sure all other clients have been configured to limit the number of CPUs being used so that each additional client you create gets 1 CPU, assuming you're only going to be running the single T4T task on that client.
-In the ProgramData folder, create a folder named BOINC_template
-Create a folder named projects in the BOINC_template folder.
-Copy the following files from the main BOINC folder to BOINC_template:
account_lhcathome2.cern.ch_test4theory.xml
remote_hosts.cfg
gui_rpc_auth.cfg
cc_config.xml (if you have one)
-Be sure the IP address of any PC, including the one you're setting up for additional clients, is in file remote_hosts.cfg
-Put the following in cc_config.xml:
<cc_config>
<log_flags>
</log_flags>
<options>
<ncpus>1</ncpus>
</options>
</cc_config>
-It can contain other settings if you want, but this is all it needs for a single T4T task.
-Folder BOINC_template is now a template for as many additional boinc clients running T4T as you wish to run.
-For the second boinc client (first additional client), in folder ProgramData, create folder boinc2.
-Copy the contents of BOINC_template to boinc2.
-For a third client (second additional), in folder ProgramData, create folder boinc3.
-Copy the contents of BOINC_template to boinc3.
-And so on for as many additional clients you want.
-Next, create folder to put your CMD files you'll use to start the additional clients.
-For the second client, create a file named boinc2.cmd in that folder.
-Edit the boinc2.cmd and add the following line:
"c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc2\ --gui_rpc_port 31417
-For additional clients, create additional cmd files, boinc3.cmd and so on.
-For each cmd file, change the boinc folder to boinc3 and so on.
-Open a Windows command prompt.
-Change dir to the folder with the cmd files.
-Run the boinc2.cmd file.
-Open your BOINC Manager.
-Select computer using the <IP address>:<port> of the new client and enter password.
-Verify it initializes the Test4Theory project, downloads a task and runs it properly.
-Change computing preferences if you want.
-If you want to stop the client, just press CTL-C in the command window and it will shut down.
-Start additional clients.
For Win7 and Win8...
-Be sure all other clients have been configured to limit the number of CPUs being used so that each additional client you create gets 1 CPU, assuming you're only going to be running the single T4T task on that client.
-In the ProgramData folder, create a folder named BOINC_template
-Create a folder named projects in the BOINC_template folder.
-Copy the following files from the main BOINC folder to BOINC_template:
account_lhcathome2.cern.ch_test4theory.xml
remote_hosts.cfg
gui_rpc_auth.cfg
cc_config.xml (if you have one)
-Be sure the IP address of any PC, including the one you're setting up for additional clients, is in file remote_hosts.cfg
-Put the following in cc_config.xml:
<cc_config>
<log_flags>
</log_flags>
<options>
<ncpus>1</ncpus>
</options>
</cc_config>
-It can contain other settings if you want, but this is all it needs for a single T4T task.
-Folder BOINC_template is now a template for as many additional boinc clients running T4T as you wish to run.
-For the second boinc client (first additional client), in folder ProgramData, create folder boinc2.
-Copy the contents of BOINC_template to boinc2.
-For a third client (second additional), in folder ProgramData, create folder boinc3.
-Copy the contents of BOINC_template to boinc3.
-And so on for as many additional clients you want.
-Next, create folder to put your CMD files you'll use to start the additional clients.
-For the second client, create a file named boinc2.cmd in that folder.
-Edit the boinc2.cmd and add the following line:
"c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc2\ --gui_rpc_port 31417
-For additional clients, create additional cmd files, boinc3.cmd and so on.
-For each cmd file, change the boinc folder to boinc3 and so on.
-Open a Windows command prompt.
-Change dir to the folder with the cmd files.
-Run the boinc2.cmd file.
-Open your BOINC Manager.
-Select computer using the <IP address>:<port> of the new client and enter password.
-Verify it initializes the Test4Theory project, downloads a task and runs it properly.
-Change computing preferences if you want.
-If you want to stop the client, just press CTL-C in the command window and it will shut down.
-Start additional clients.
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#18 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Look at The Scottish Boinc Team! One day in and we're in 1st, with just 4 of us on the board so far! Let's see if we can hold it. :bounce: Calling all Scots!
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#19 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Back down to 2nd, but for a CPU only project that is awesome.
As are you scole, above us all and up to 4th!
I was pleased with myself at getting a WU running on most of my machines but you've got one on each core!
I'll have a go at this myself see if I can get it to run.
As are you scole, above us all and up to 4th!
I was pleased with myself at getting a WU running on most of my machines but you've got one on each core!
I'll have a go at this myself see if I can get it to run.
Willie the Megacruncher
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#20 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
GEEZ! Important note. I messed the file list for the template. Should have been...
account_lhcathome2.cern.ch_test4theory.xml
NOT master_lhcathome2.cern.ch_test4theory.xml.
Hope I didn't cause a bunch of grief and wasted time.
account_lhcathome2.cern.ch_test4theory.xml
NOT master_lhcathome2.cern.ch_test4theory.xml.
Hope I didn't cause a bunch of grief and wasted time.
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#21 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
We're back to #2. BOINC.Italy is not out of reach though.
I'm currently sitting at #2. We TSBT pack a punch! Is the old guard on TSBT not up to catching me?
I'm currently sitting at #2. We TSBT pack a punch! Is the old guard on TSBT not up to catching me?
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#22 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
If you mean me, then the answer is almost certainly not!scole250 wrote:Is the old guard on TSBT not up to catching me?
I'm languishing well behind you in 9th and even the mighty Derv can only manage 6th.
I'll do my best though to help win the two horse race developing between us & the Italians. Less than 2 results in it! :)
Willie the Megacruncher
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#23 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I'm throwing down the gauntlet to the whole team. We can win this challenage! :)
- Janos (retired)
- Still a Newbie
- Posts: 1919
- Joined: Thu Feb 23, 2012 8:58 am
- Location: Aberdeenshire, Scotland
#24 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
"Happiness can be defined as: a geek with non-work related code to write, no distractions and no deadline." - Janos
#25 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Jeez, go away for 2 days with the kid, come home to my network down for 3 of the machines running T4T and way down the order . When I get home tonight will have to get the dual quad zeons installed and have a go at this multiple clients malarky
What's the deal with the ip address ? Do I have to specify an ip address for each multiple client I set up on the same machine ?
What's the deal with the ip address ? Do I have to specify an ip address for each multiple client I set up on the same machine ?
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#26 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
No separate IP address, just a different port for the manager app to connect to. Port 31416 is the default port your current client uses. It was suggested to just increment the others as you go to make it easy to remember. Port 31417 for the second client, 31418 for the 3rd and so on.Alez wrote:What's the deal with the ip address ? Do I have to specify an ip address for each multiple client I set up on the same machine ?
#27 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
so i only have to change this line
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc2\ --gui_rpc_port 31417
and 3rd becomes
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc2\ --gui_rpc_port 31418
etc.
then when i connect with boinc to each client
-Open your BOINC Manager.
-Select computer using the <IP address>:<port> of the new client and enter password.
the ip address is the same , only the port changes.
Where do I set or find the password ?
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc2\ --gui_rpc_port 31417
and 3rd becomes
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc2\ --gui_rpc_port 31418
etc.
then when i connect with boinc to each client
-Open your BOINC Manager.
-Select computer using the <IP address>:<port> of the new client and enter password.
the ip address is the same , only the port changes.
Where do I set or find the password ?
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#28 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Third becomes boinc3...
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc3\ --gui_rpc_port 31418
Fourth becomes folder boinc4, port 31419...
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc4\ --gui_rpc_port 31419
You have to copy the contents of the template folder into each boinc folder.
Also, under computing preferences, set While processor usage is less than to 0. Each client doesn't know about the others and counts their cpu% as something other than itself. If it's set to 50%, that client won't do anything if other clients are using 50% of cpu.
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc3\ --gui_rpc_port 31418
Fourth becomes folder boinc4, port 31419...
c:\program files\boinc\boinc" --allow_multiple_clients --dir c:\programdata\boinc4\ --gui_rpc_port 31419
You have to copy the contents of the template folder into each boinc folder.
Also, under computing preferences, set While processor usage is less than to 0. Each client doesn't know about the others and counts their cpu% as something other than itself. If it's set to 50%, that client won't do anything if other clients are using 50% of cpu.
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#29 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Another question. :)
I couldn't find remote_hosts.cfg but soldiered on regardless. So far it's not working out well on the first machine I tried it on with all my virtual machines becoming inaccessible after 8 hours of seeming to run well enough.
Do I actually need this file & if so what do I put in it?
I couldn't find remote_hosts.cfg but soldiered on regardless. So far it's not working out well on the first machine I tried it on with all my virtual machines becoming inaccessible after 8 hours of seeming to run well enough.
Do I actually need this file & if so what do I put in it?
Willie the Megacruncher
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#30 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
The remote_hosts.cfg file contains the IP addresses of PCs whose BOINC manager is permitted to connect to that client. If the manager will connect to the client, I guess it's not important. This would only work if connecting to a client on the same PC.
I have a PC that had virtual hosts that didn't act right. The % complete would never update. Not sure those tasks were working correctly. I flushed everything, started from scratch and ran as Administrator. Not sure if that was the solution, but it seems to be working now. Looking back, the % complete may not have been updating because I may have had the run While processor usage less than % set where it wasn't actually running.
I think the thing that tells you it's running correctly is the task % complete progresses and you observe the total cpu load reflect the number of tasks running. I had 6 WUs running but cpu at 40% until realized the run While processor usage less than setting was causing tasks to not actually run.
I have a PC that had virtual hosts that didn't act right. The % complete would never update. Not sure those tasks were working correctly. I flushed everything, started from scratch and ran as Administrator. Not sure if that was the solution, but it seems to be working now. Looking back, the % complete may not have been updating because I may have had the run While processor usage less than % set where it wasn't actually running.
I think the thing that tells you it's running correctly is the task % complete progresses and you observe the total cpu load reflect the number of tasks running. I had 6 WUs running but cpu at 40% until realized the run While processor usage less than setting was causing tasks to not actually run.
#31 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I don't have the remote host file either .
So far I haven't figured out how to attach to the client once I have got it running under the cmd line Probably doing something stupid but where do I connect ? I have the option of select computer or launch another boinc manager but both just ask for a host name and password ?
and I just crashed my vbox and lost 8 hours
So far I haven't figured out how to attach to the client once I have got it running under the cmd line Probably doing something stupid but where do I connect ? I have the option of select computer or launch another boinc manager but both just ask for a host name and password ?
and I just crashed my vbox and lost 8 hours
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#32 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
To connect to the client, select computer and under host name enter <ip address>:<port>. Even if it's on the same PC you must enter the IP address and port. Using the PC name hasn't worked for me. Had to use the IP address. If connecting from a different PC, you of course need the manager PC IP address in the remote_hosts.cfg.
Another note. You can over allocate our CPUs. Multiple clients don't know about each other, so if you can run 8 threads, you can run actually 9 (or more) clients, each thinking it's using (assigned) a CPU, but really just get a share of what's available.
Another note. You can over allocate our CPUs. Multiple clients don't know about each other, so if you can run 8 threads, you can run actually 9 (or more) clients, each thinking it's using (assigned) a CPU, but really just get a share of what's available.
#33 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I don't have an ip address as my address are all dynamic for 21 different devices assigned by my network switch and have no remote host file. Tried to connect using computer name localhost:31417 ( which should work )but that then connects me to the same host as my primary ie localhost:31416. If I then try and switch back my system almost locks up as I seem to be running 2 instances of the same host.
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#34 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Ping the machine name to get the IP address of other PC or to get IP of current PC, run ipconfig at the command prompt.
#35 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
pinging the machine gives me the standard 225.225.225.0 the same as localhost. From the command line I can get a new host but as it has the standard ' do not run if usage more than 25%' etc. it does not download a unit and I can't connect to the new host to change the presets.
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#36 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
255.255.225.0 is the subnet mask. Is there an address in the list like 192.168..x.x , other than the gateway? If so, that's the address.
#37 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I get an ipv4 address of 192.168.56.1 which I can't use either. I also get an vbox address which will not connect either. Connects then disconnects. Will have to think about this later in the morning. Obviously I have something set up wrong ( I guess)
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#38 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
You're trying to connect to the actual virtual box. You don't need to. The vboxwrapper app is the only thing that interacts with it. Connect to the host pc ethernet adapter local area connection address and use the port also, like 192.168.1.20:31417
#39 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Yes I was trying to connect to vbox as that made sense at the time. Thanks for pointing me in the right direction.
What a faf but finally got it working, but slightly different and modified.
Here's how.
make directory boinc2 in c:/ProgramData/
copy account_lhcathome2.cern.ch_test4theory.xml and gui_rpc_auth.cfg from c:/ProgramData/BOINC to c:/ProgramData/boinc2
Make cc_config.xml file with notepad and save in c:/ProgramData/boinc2
insert the following into the cc_config.xml
<cc_config>
<options>
<ncpus>1</ncpus>
<data_dir>C:/ProgramData/boinc2</data_dir>
</options>
</cc_config>
Make boinc2.cmd with notepad and save where ever is handy.
insert the following into boinc2.cmd
"c:\Program Files\BOINC\boinc.exe" --allow_multiple_clients --dir c:\ProgramData\boinc2\ --gui_rpc_port 31417 --detach
note different spelling from previous postings
in your main boinc client reduce cpu usage by 1 ( set ncpus in config file or simply reduce core percentage )
double click on boinc2.cmd cmd window will open , run then close.
in boinc client - advanced - select computer.
for host name insert 127.0.0.1:41317
open gui_rpc_auth.cfg with notepad, copy the long string of letters and numbers and paste this into the password.
you are now running 2 T4T units. set the preferences you wish ( will probably be suspended at first as use computer whilst busy will be unselected ) You will also need to change suspend if busy from 25 to 0 and I think specify a disk size.
to run a third app do the same changing data dir and port to 41318 in boinc3.cmd and data dir to boinc3 in cc_config.xml
to return to your main client - advanced-select computer host name is 127.0.0.1:41316 or localhost:41316
password if required is the same as already pasted from gui_rpc_auth.cfg
What a faf but finally got it working, but slightly different and modified.
Here's how.
make directory boinc2 in c:/ProgramData/
copy account_lhcathome2.cern.ch_test4theory.xml and gui_rpc_auth.cfg from c:/ProgramData/BOINC to c:/ProgramData/boinc2
Make cc_config.xml file with notepad and save in c:/ProgramData/boinc2
insert the following into the cc_config.xml
<cc_config>
<options>
<ncpus>1</ncpus>
<data_dir>C:/ProgramData/boinc2</data_dir>
</options>
</cc_config>
Make boinc2.cmd with notepad and save where ever is handy.
insert the following into boinc2.cmd
"c:\Program Files\BOINC\boinc.exe" --allow_multiple_clients --dir c:\ProgramData\boinc2\ --gui_rpc_port 31417 --detach
note different spelling from previous postings
in your main boinc client reduce cpu usage by 1 ( set ncpus in config file or simply reduce core percentage )
double click on boinc2.cmd cmd window will open , run then close.
in boinc client - advanced - select computer.
for host name insert 127.0.0.1:41317
open gui_rpc_auth.cfg with notepad, copy the long string of letters and numbers and paste this into the password.
you are now running 2 T4T units. set the preferences you wish ( will probably be suspended at first as use computer whilst busy will be unselected ) You will also need to change suspend if busy from 25 to 0 and I think specify a disk size.
to run a third app do the same changing data dir and port to 41318 in boinc3.cmd and data dir to boinc3 in cc_config.xml
to return to your main client - advanced-select computer host name is 127.0.0.1:41316 or localhost:41316
password if required is the same as already pasted from gui_rpc_auth.cfg
The best form of help from above is a sniper on the rooftop....
#40 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
We are still 2nd, only 4.4k behind though, a gap that we can bridge in the next 3 days
Scole has grabbed 1st place Way to go
I'm sure there is a team victory in here too
Scole has grabbed 1st place Way to go
I'm sure there is a team victory in here too
The best form of help from above is a sniper on the rooftop....
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#41 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Aye! Let's go lads and lassies! I'm going to give up my last 2 cores on collatz for a few days. Give till it hurts!
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#42 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Don't sacrifice Collatz for it.
But that aside we could well win this! All the faffing about has cost me a few WUs but finally the benefits are eclipsing the costs. In the last 2 hrs I've gone from 10th to 7th overtaking Derv in the process. If Alez's many virtual machines last the night then he too will roar up the ranks.
Latest is that scole is back on top & TSBT is just 2.9 K behind the Italians.
I predict we will be ahead this time tomorrow!
But that aside we could well win this! All the faffing about has cost me a few WUs but finally the benefits are eclipsing the costs. In the last 2 hrs I've gone from 10th to 7th overtaking Derv in the process. If Alez's many virtual machines last the night then he too will roar up the ranks.
Latest is that scole is back on top & TSBT is just 2.9 K behind the Italians.
I predict we will be ahead this time tomorrow!
Willie the Megacruncher
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#43 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
And TSBT are a couple of K in front again. :bounce:
Encouraging but lots of race yet to run. Go, go & thrice go! :)
Encouraging but lots of race yet to run. Go, go & thrice go! :)
Willie the Megacruncher
#44 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Me ? Many virtual machines ? Am I that transparent ?Megacruncher wrote:If Alez's many virtual machines last the night then he too will roar up the ranks.
Most of them made it through the night..... though some didn't Very frustrating when things crash and you loose 18 hrs of crunching to ' computation error '
No idea how much this will help but hopefully I should start dumping credit from 4pm ish.
Italy sneaking away from us again
The best form of help from above is a sniper on the rooftop....
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#45 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Only to users of the internet. :)Alez wrote:Me ? Many virtual machines ? Am I that transparent ?Megacruncher wrote:If Alez's many virtual machines last the night then he too will roar up the ranks.
http://lhcathome2.cern.ch/test4theory/h ... erid=28895
Most of your machines have gone the distance & added a lot to your total in the last 24hrs . Indeed you have overtaken me. Once I've uploaded a heap of Wus , hopefully in the next hour or so, the situation should look a little better and I'll see if I can fit in a few more VMs. As it stands a 3.4K disadvantage is easily overturned in the 50 hours remaining to us.Alez wrote: No idea how much this will help but hopefully I should start dumping credit from 4pm ish.
Italy sneaking away from us again
Willie the Megacruncher
#46 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Eeek, big brother is watching
I lost 3 vm's to crashes and had another 4 lock up so they are not lost only 'delayed'. Not sure if it is temperature related or running out of memory perhaps. If i can keep them up for the next two days that should keep us with the Italians allowing you and Scole to kick us past them
Crack on guys
I lost 3 vm's to crashes and had another 4 lock up so they are not lost only 'delayed'. Not sure if it is temperature related or running out of memory perhaps. If i can keep them up for the next two days that should keep us with the Italians allowing you and Scole to kick us past them
Crack on guys
The best form of help from above is a sniper on the rooftop....
- Megacruncher
- G.L.S.B.
- Posts: 4700
- Joined: Mon May 29, 2006 11:33 pm
- Location: Edinburgh, Scotland
- Contact:
#47 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Yup it's a labour intensive comp this one, keeping VM and Boinc Manager and multiple instances of the Boinc Client all playing nicely together!
Anyhow we are back in front albeit only by about 700cr. :bounce:
We usually do well overnight so let's hope we have a useful lead by the morning!
Anyhow we are back in front albeit only by about 700cr. :bounce:
We usually do well overnight so let's hope we have a useful lead by the morning!
Willie the Megacruncher
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#48 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
No kidding! I'm exhausted, but I can smell a 1st. I hope that's what that is.Megacruncher wrote:Yup it's a labour intensive comp this one, keeping VM and Boinc Manager and multiple instances of the Boinc Client all playing nicely together!
- scole of TSBT
- Boinc Major General
- Posts: 5982
- Joined: Mon Feb 03, 2014 2:38 pm
- Location: Goldsboro, (Eastern) North Carolina, USA
#49 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
Time for some horse trading. Janos, Paul and any other TSBT member, if you will put your cores on this challenage for the next 2 days, I'll give the same number of cores back for a week on project(s) of your choice. It's for a top finish!
- Janos (retired)
- Still a Newbie
- Posts: 1919
- Joined: Thu Feb 23, 2012 8:58 am
- Location: Aberdeenshire, Scotland
#50 Re: 23/5-30/5 CERN's 60 years of science LHC T4T@home
I'm afraid I am rather limited in what I can do at the moment. Major config changes from afar are not so easy.
"Happiness can be defined as: a geek with non-work related code to write, no distractions and no deadline." - Janos