sudo nm-connection-editorQuote:
Originally Posted by nanoprobe
Printable View
sudo nm-connection-editorQuote:
Originally Posted by nanoprobe
Question for you Linux guys. I am running a Ubuntu 10.10 pc on a wireless network. The wireless network went down for a few minutes yesterday and 2 days worth of WU's showed computation errors. I figured even without an internet connection it would countinue to crunch but it errored them all out. Is there something I can do to prevent this from happening again?
It was not the wireless going down that caused it. They crunch just fine without internet, just can not send or receive new or completed ones. Whatever took down wireless may have caused the failures. In other words the wireless going down may be a symptom of another problem. My internet went down for several hours, wired and wireless on 7 crunchers and 4 projects. No failures or errors.Quote:
Originally Posted by Moo
If you're running with an overclocked PCI bus and it's become a little flakey then all manner of weirdness will ensue. Alternatively if running too hot (even at stock clocks) has caused memory corruption then there's no telling what might happen.
The PCI bus is at 100 and the PC is in a room that is 50 degrees fahrenheit. The processor is lucky to see 40 degrees celsius. Once the wireless was back up an running it uploaded everything and downloaded new WU's and went back to crunching. It has been running now for about 36 hours without a hiccup. The only thing I can think to do is cut back my WU's to .10 of a day and kill the wireless and see what happens.Quote:
Originally Posted by D_A
I need some help from the guru's. Just installed a second E5472 on my asus board. With one processor, processor was crunching at 100% with. Few dips. Installed 2nd chip and seems ubuntu 10.10 is throttling like crazy. My other dual processor E5420 on a asus board runs 100% with very little dipping. Only difference between boards is one has 800 fsb, the other is 677 fsb. This has me baffled as to why one is barely throttling and the other is throttling real bad. Any suggestions?
If you are using the GUI System Monitor it will really screw things up, it uses a much highier priority than anything else and WCG use just goes to nothing. Open a terminal window and use top. See how things are then and let us know.
cheers for that tip poppageek :up:
Haven't read the thread , but CEP2 is VERY disk intensive.
I only run 1 CEP2 WU at the time on my Hexacores,
Maybe that's why it's not running smooth?
I just switched from fighting aids. Did the same thing. What really ticks me off is my other rig,(dual E5420 on asus board 677 FSB) stays pretty much pegged at 100% crunching aids. Doesn't make sence. Both running ubuntu 10.10.
CEP2 is both disk I/O and memory intensive. VERY intensive for disk I/O. If the one throttling has a slower HDD installed you will notice it running CEP2 exclusively. I wouldn't mind betting that memory speed might actually make a difference with this project.
I notice you're using most of your installed RAM plus some swap space. Perhaps you could try adding some more RAM and see if that helps. Also, you don't (certainly shouldn't) need to leave boincmgr running. It's just an interface, not the client itself.
As for actual throttling try entering this in a terminal and then rebooting:
It will remove the ondemand CPU throttling system from all run levels and set the PCU to run at 100% speed constantly. Be warned that the system MIGHT run a little warmer after doing this.Code:sudo update-rc.d -f ondemand remove
Jaco is right CEP2 will slow things down running that many. VERY disk intensive. WUs are waiting for IO. When not needed close Boinc Manager, that is what is causing Xorg to use so much. Why there are 9 CEP apps running I do not know. The one getting no CPU time has a different name.
That one might have been pre-empted for some reason. Possibly about to finish another one? Hard to tell.
I had a machine running 8 CEP2 units. It was a dedicated cruncher. It would error out all of it's WUs within a few seconds of each other. After switching to using a ANS-9010 ramdrive the issue disappeared. There seems to be some kind of limit in Boinc for disk I/O. If things get too crazy the whole house of cards comes crashing down.
I found the old thread and on my SR-2 just starting 16 threads created 2.5GB of data written in a few seconds. An hour later 27GB had been written to the drive. At 36 hours 1.5TB had been written to the drive!
See http://www.xtremesystems.org/forums/...7&postcount=29
I'd be very careful running those units exclusively on a system with a SSD.
Went ahead and reduced the number of wu's to 4 per day on both my harpertown crunchers. My fans a ramping up pretty good doing 100% cep2. I'll run that command in the morning after it dumps overnight. One big drawback with cep2 is it clogs the hell out of my Internet connection when uploading. Was hoping a 7200 rpm drive would keep up but I guess 8 running at the same time is a little to much. I swore I wouldn't come back to this project but hate looking at the green badge knowing I can ruff it a few months to get 2 years done. Thought I read they optimized the wu's a little better but looks to still be a royal PITA. Oh well. Thanks folks. Appreciate the help.
I heard the same, but "a little better" might only be a few %. I limit my machines to 3 because of the upload size. I'm not running any right now, but single core machines (at least mine anyway) HATED CEP2 units. 100% error rate. Never got it sorted, either. All my multicore machines run them with no trouble, aside for the big uploads choking the network. I did find that running a network proxy helps that as the units load to it first and then up, but it's not a huge benefit.
Someone over at WCG turned me on to this web site for possible help with this.
http://www.experts-exchange.com/OS/L...peed-Tips.html
Tried a few things but it still throttled. Then I switched to CEP2. Probably a bad idea until I sorted this out. Anyone think I should try a reinstall to see if this would help?
Sometimes I am a little dense so if I am missing something help me out.
You have 8 WUs running. 4 @ 100%. If you take the % for Xorg and BoincManager and add those numbers to the remaining 4 you are 99-100% on the other 4. :shrug:
This is a 8 core right machine right? You do not have to run BoincManger, I do not. Xorg will drop to low single digits if no X windows are open. You could also just not run X.
Ya know, I plain forgot to check that. Out of a possible 800% (8 cores at 100%) top is showing a sum total of 797%, breaking that down that means the machine is at 99.6% total CPU utilisation. That's actually pretty good! All the tweaks in the world will not gain you any more than that last 0.4%, though ditching X and boincmgr will gain you the equivalent of 20% of one CPU (roughly) by removing competing processes.Quote:
Originally Posted by PoppaGeek
Woke up this morning to find error's up the kazoo. Reading the log shows right after it connected at 1am to do it's upload, what wasnt crunched went to errors. Decided to reduce even more to just 2 per machine. To much a royal PITA. Went ahead and assigned each core to performance instead of using commands in terminal. Will see what happens. Thanks again.
BTW, this is a 8 core machine. 2 E5472's.
Figured it out, by accident. Changed in the bios to not allow speed stepping, which locks the processor to it's highest speed. BINGO, am maxed at 100% at each core with minor dips periodically.Quote:
Originally Posted by bearcatrp
FYI- The package command for Ubuntu now installs 6.10.58 of Boinc.
sudo apt-get install boinc-client boinc-manager or
sudo apt-get update boinc-client boinc-manager
I discovered I'd forgotten to disable this on one of my Sossaman boards only a few days ago. :shakes: This is something I normally look out for.Quote:
Originally Posted by bearcatrp
alright so what do i do when i click on Attach To Project and it won't?
help...
Linux hates you. You must die. :slash:
Would help with more info. Error message? Login ok? No projects showing? What? Is this Dotsch? How was Boinc installed?
Make sure the client is running.
sudo /etc/init.d/boinc-client restart
omg information overload. It's Ubuntu 10 i believe, installed from the boinc website. Everything was crunching just fine when i shut the machine down but now when i try to start Boinc manager it won't connect to the local host (whatever that means) this isn't the first time this has happened but i can't remember how i fixed it before. i don't know what else to say. I hit "Connect to Project or Account Manager" and i get an error. nothing more, just says "Error".
Are you using the script in the BOINC directory to start the manager (run_manager) or are you just clicking on/linking to the binary file? (boincmgr) If you do the latter it will misbehave.
Using apt-get to install sets up the client to start on boot. Installing from the download at Berkly does not so you have to start it manually or setup the startup on boot yourself. That's why it was recommended to install from the apt-get command.
I'm starting it by opening the folder on my desktop and clicking on the file called boincmgr (i think)Quote:
Originally Posted by D_A
But the version in the repos is way outdated so that's why i went with the one on berkeley's siteQuote:
Originally Posted by PoppaGeek
should i uninstall the version i have and start over?
Leave what you have installed. There's no need to remove it. When you extracted the original file it would have told you to use the run_manager and run_client scripts to kick things over, NOT the boincmgr or boinc files. The others set up the environment variables so the programs can find everything. Just do it that way and you'll be fine.Quote:
Originally Posted by shoota
Attachment 114052
Attachment 114051
crap. i think it's @*@&#^$. The biggest problem I have now is that I can't connect to the internet. Neither wired nor wireless. Between this board and linux I'm gonna pull my hair out. I don't even know where to start in getting the internet working again. Any suggestions in network troubleshooting?
Open a terminal and enterand post the output. You might well have just moved a network cable or something.Code:ifconfig
So... I can't believe I can't figure this out, but I'm currently at a loss.
I'm trying to install Boinc remotely on a computer I won't have physical access to for two weeks. I used Cygwin to access it and log in, but once I pull it up I can't figure out where I am. If i ls it gives me nothing and .. returns -bash ..: command not found.
The terminal says [MyLogin@Computer ~]$
If it makes a difference it's running CentOS.
I know I'm being an idiot... but yeah.
Edit: wow... forgot the cd for the ..
I might just need a new brain...
Edit 2: So where do I DL and install it to. (This one I dont feel so bad about, never installed anything in linux... which I actually kinda feel bad about. :p:)Here are my options.
Ah... so now its up and running... but it's only recognizing one processor. :rolleyes:
I just emailed a... I guess he's a mentor?... who knows more about this stuff than I do so we'll see what he says.
Anyone know how to get a screenshot of all the threads in CentOS? Cuz that's a pic I wanna see. :rofl:
Yet another question for a linux guru when they get around to it...
How can I install, run and control BOINC if I cannot access anything other than a shell?
By using the command line. Everything you can do with the manager can be done with CLI commands as well. Google boinccmd and you should get a list pretty quickly.
I'm thinking about switching to command line on my Arimas. Then uninstall the GUI. I've had weird issues with it. And I don't even touch them except to update WU or reboot for patches.
So Linux gurus; I have been running mainly Ubuntu desktop on my non Windows machines. I have used all versions 8.04 thru 10.10 always the desktop version for ease of installation. I don't use most of them for anything but crunching. I am going to get rid of a lot of P4's this summer most of them are XP machines, old P4 Dells, the electricity is eating me up for the PPD output and the AC to cool the house in the summer. I will probably be getting a couple Sandy Bridges to replace them all but I may want to wait until the P68 MBs are out so I can run without video cards. Now that Ubuntu11.04 is out what crunching improvements could I expect by going to the server version versus the desktop or a lightweight desktop like Lubuntu? More PPD or faster WU processing? Most all machines seem to not have a memory or HD shortage issues at present. These are Intel I5's and I7's and AMD X6's and Athlon II X4's and 3 Sossaman servers that run Linux now. Most all single cores are going away both AMD & Intel. Most are on wired Ethernet. I am not afraid of a CLI interface but haven't had the best of luck to date with putting it all together with Linux to make it do what I want and have had to rely on the GUI's to help get things running. :confused:Quote:
Originally Posted by PoppaGeek
Running without a GUI will definitely net you both more points and more work completed, the only real question is, how much. Even when you don't have any memory or storage issues, any GUI will be stealing CPU cycles just to keep itself running. Any other additional services, like Bluetooth, will do the same thing, plus the more stuff you have on the system the more it will keep wanting to update so running a server install with only those daemons running you absolutely need will give you the best output from your machine. We're still only talking, MAX, 10% difference if you're lucky, I'd guess.
It's not that challenging to maintain a headless machine via SSH once it's up and running, plus BOINC administration can be done remotely from a desktop machine easily enough. Networking an be a pain from the CLI, but it's no trouble if you set it up at installation.
Need some help on 2 issues.
1. Not sure how this happened but I sold my harpertown rig but kept the hard drive for my next build. I installed it on another PC I was testing and it's in command line mode, not the GUI. How do I revert it back?
2. I have a ATI 6950. Running ubuntu 10.10 64. ATI catalyst is v2. I have to download the latest version manually but can't remember the command to install it once I put the file on my desktop.
Appreciate any help.
q1: You've probably got a different video card in the new system. Hold down the left shift key while it's booting to get into the GRUB menu, then select the recovery option which should be the second one, then in the next menu select the one that refers to reconfiguring the X-server or video configuration or something like that and it will do it itself.
q2: Downloaded the latest version of what?
ATI catalyst. It's at v11.6
http://support.amd.com/us/gpudownloa...2&lang=English
What I have now doesn't support my 6950.
See if this helps: https://help.ubuntu.com/community/BinaryDriverHowto/ATI
Thanks for the help. I will give it a shot.
Here is the output from my system below.
Its running Ubuntu server, command line only
Attachment 118222
i'm currently working on putting toghetter a non stop crunching server based on what i could get for almost nothing Tyan Thunder K8W (http://www.tyan.com/archive/products...hunderk8w.html)
and 2 opterons 240 (1.4 Ghz) i'm trying to get 2 opterons 280 (2.4 ghz dual core)
is it best to run this on linux for its stability?? or would u guys advice on using windows server...
maybe this is the wrong place to ask but still
i use linux basically because it's free. 4 of my 5 crunchers are linux
ok and installing the boinc manager is just as easy ?? :P
Depending on the distribution you choose the installation can vary but it's pretty simple. Just don't use FreeBSD or Solaris because they complicate things more than you need for this. I think Ubuntu, Debian or something based on them be the most popular here.
yeah i worked with ubuntu before so i gues that will do the trick (apt get is realy nice :P)
It's pretty nice, I'm quite fond of it as well.
There's a whole thread on getting BOINC up and running on Ubuntu, both 32 and 64bit (recommended if your hardware will take it) in this part of the forum.
Worked like a charm. Thanks. Haven't had time to try the second question yet.Quote:
Originally Posted by D_A
got it working linux 64 bit with 2 opteron 242 (1.6 ghz) procs :)
Got the updated catalyst up and running. Pretty simple after all. Thanks again.
Have another issue that popped up. Looks like one of my sticks of ram is out. Is there a program or command to see which slot has the dead stick? I have 12gb installed but only 10gb being reported. Already swapped them around to see if I had a loose one.
Try: sudo dmidecode -t 6
& dmidecode -t 17
Worked like a charm. Thanks.
Thought I would resurrect this thread due to recent increase in Linux interest.
Going thru this thread checking if links are still good and info up to date. I have not seen a command mentioned I often find useful.
file <filename>
Will tell you the type of file it is.
Quote:
poppageek@8356x2deb:~/BOINC$ file boinc
boinc: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.24, BuildID[sha1]=0x715e83716479c4a94b31cb45948d3e7ce245c02a, stripped
Quote:
poppageek@8356x2deb:~/BOINC$ file binstall.sh
binstall.sh: ASCII text
Quote:
poppageek@8356x2deb:~/BOINC$ file ca-bundle.crt
ca-bundle.crt: ASCII English text
Updating to latest Boinc on Berkely site, Linux
I wanted to update Boinc to the latested stable one at Berkely so I could use the app_config.xml file as the latest Boinc from the Ubuntu repositories was 7.0.27 and I needed 7.0.40 at least to use it. While it is still fresh in my mind I thought I would document how I did it. I know D_A has given the directions to do it like this but I could not find it in the forums, so I thought it might be usefull with the renewed interest in Linux thanks to MMs adventures.
What you type is in bold text.
Install Boinc from repositories, if not already installed. If it is installed skip this part.
sudo app-get install boinc-client boinc-manager
Now stop boinc from running. When you installed it Boinc was started.
close Boinc Manager if open.
sudo service boinc-client stop
Download latest Boinc from Berkely to your $HOME directory.
In a Terminal window:
type cd and hit enter, this will put you in your home directory
NOTE: you do not have to type the entire file name. Type boinc_ and hit tab, it should finish the filename for you. If not type more of the name and hit tab until it does.
sudo chmod a+x boinc_7.0.65_x86_64-pc-linux-gnu.sh or what version you downloaded
sudo sh boinc_7.0.65_x86_64-pc-linux-gnu.sh again substitute the name of the file you downloaded
now cd BOINC the directory name should be all caps.
sudo cp boinc* /usr/bin/
sudo service boinc-client start
Setting up scheduled BOINC client updates with cron (Linux)
Good tutorial for setting up a crontab to run jobs (tasks or scripts) at regular intervals.
Most of this has been covered elsewhere but it can really make the command easier to live with.
Getting around with the command line and editing it.
$HOME is users home directory. /home/username/ Mine is /home/poppageek/
pwd is Present Working Directory.
Bold is what you type at command line.
cd change directory to users $HOME
cd .. Go back one directory.
cd - Go back to directory I was previously in.
Quote:
poppageek@8356x2deb:/bin$ cd /etc
poppageek@8356x2deb:/etc$ cd -
/bin
poppageek@8356x2deb:/bin$
Hitting the [TAB] key will finish a filename or a directory name for you.Quote:
poppageek@8356x2deb:/bin$ pwd
/bin
poppageek@8356x2deb:/bin$ cd /var/lib/boinc-client/projects/
poppageek@8356x2deb:/var/lib/boinc-client/projects$ cd -
/bin
cd /var/lib/boin[TAB] will show cd /var/lib/boinc-client hit [ENTER]
chmod +x boinc_[TAB] becomes chmod+x boinc_7.0.65_x86_64-pc-linux-gnu.sh (makes boinc_7.0.65_x86_64-pc-linux-gnu.sh executable.)
If you hit [TAB] and nothing happens hit it again. A directory listing of all files and directories that begin with the letters you typed will be listed.
All directories that start with b in /etc are listed.Quote:
poppageek@8356x2deb:~$ cd /etc/b[TAB][TAB]
bash_completion.d/ bluetooth/ boinc-client/ byobu/
Type o and hit [TAB] then [ENTER] and you will be in /etc/boinc-client Since there is only one directory in /etc/ that starts with bo it knows how to finish what you are typing.
ls -la will list a directory contents including the hidden files. Hidden files start with a . period. As .viminfo
You can use arrow keys to edit a command.
Say you typed ls -la /var/lib/boinc-cient and hit [ENTER].
Hit [UP ARROW] key and then [LEFT ARROW] key to add l to client. Use [RIGHT ARROW] key to end of line. Hit [ENTER]
[BACK SPACE] and [DELETE] keys work as expected.
The command line history remembers more than your last command. Keep hitting [UP ARROW] until you find the command you want to repeat or edit.
A tool that I became VERY fond of a little while back was cssh, or ClusterSSH as the project is called. It allows the user to control multiple systems via command line at once. In practice, once it's set up you get a little window you type your commands into and cssh mirrors that command across multiple target machine.
The up side is this can save you a TON of typing if you are administering a bunch of identical installs, such as headless Linux crunchers.
The DOWN side is that if you screw up one command you can bork huge banks of machines with a single key stroke. This is a "Spiderman" tool, great power coming with greater responsibility.
For those wanting the power, and willing to accept the consequences, here's how to get a basic setup running.
In Ubuntu or Mint
enter your password and accept the prerequisites.Code:sudo apt-get install cssh
Once it's installed you can run it two ways (well ... several really). Either enter cssh and the IPs or hostnames (if you have a name server on your network) of all the machines you want to handle, or you can edit the "clusters" file.
This file, /etc/clusters, is owned by root so you have to use sudo or gksudo to edit it ... or su to root.
The structure of this file is simple enough
Each cluster entry is a tag followed by the username (if needed) and either the host name or IP
eg
Home = michael@192.168.1.2
the username is only required if you will be logging into an account with a different name than the one you're on.
In my case that means I can use lines like:
Home = 192.168.1.2 192.168.1.5 192.168.1.6 192.168.1.7 192.168.1.8 192.168.1.9
This ALSO requires you to have the same password for that account on every listed node. If you don't want to do that you can use a key file system and not have to enter any passwords at all, but that's a whole other walk-through.
In the current example, I would run this by enteringinto a terminal, waiting for all the windows to finish connecting, and then entering my password (for those machines). After that just enter your commands and watch in delight as they are repeated across the field of open terminal windows in front of you ... or horror as you realise you just wiped the BOINC folder off all of them instead of only one. Like I said this is not a Fisher Price tool and comes with NO safety guards, but it can save you a huge amount of typing if you're administering many identical machines.Code:cssh Home
http://img600.imageshack.us/img600/3014/rcya.png
Uploaded with ImageShack.us
Well I have got a bunch of the little Raspberry PI's so I might give this as a Project for the GEEKS in the class, install WCG onto a Raspberry PI.
Don't think it will add many points but it will be interesting to see if it can be done!
You would have to run android on those pies to get WCG tasks
Attachment 131736Quote:
Originally Posted by Evantaur
Years ago I had a FreeBSD cluster doing Distributed.net crunching. They were all Pentium Pros, 6 IIRC and a dual pentium 166 OC'd to 200 as the master that cached all the incoming and out going work units and assigned them to the nodes. Used a program called clusterit, in the FreeBSD ports, to control all the machines. Issue one command on command line and it is done on all. Obviously pretty handy and saved ALOT of repetition. All the FreeBSD machines were in a small bedroom, nothing else in there, and I would telnet in to the master node from my Red Hat box in the living room. There were only 3 machines in the Living room. Now you know why I am single.......
Glad to see something similar for Linux.
Albert@home has Raspbian work units.
Raspberry Pi BOINC Project
Albert@home is Einstien@home alpha test program. No real credit it is just for testing apps and work units.
I used this procedure to upgrade thinking it would give me the option in boinc manager to use cpus at 100% but it's not there, system monitor shows cpus cycling, how can I force 100% in Ubuntu?Quote:
Originally Posted by PoppaGeek
Tyan S4985 and (4) 8347HE's
The option is there, as far as BOINC can control it, but you need to be in the advanced view.Quote:
Originally Posted by hendermd
Click through:
View > Advanced View
In the new view click:
Tools > Computing Preferences
In the new window that opens click the "processor usage" tab (the far left one), then down the bottom are the CPU usage settings.
By default BOINC only uses 60% of CPU capacity. If you want more you need to change it yourself either view your My Grid page at WCG (which will affect all the machines using that profile) or as I described for the local machine only.
Do not use System Monitor. It runs at a higher priority than Boinc. Running it throttles Boinc. Open a terminal window and type in top it will show you an accurate reading. Press q to quit.
The other option is that the system is engaging thermal throttling, which is controlled by the BIOS. If you are confident your system cooling is adequate you can disable it there, but that is YOUR responsibility if you do.
Several years ago if you opened System Manager while running Boinc work units the CPU graphs went wild and CPU% would jump wildly. That does not seem to be the case now. I have checked it on the 2 Ubuntu boxes I have a GUI on and it acts fine. Some things improve with age.
Feel free to use System Manager with reckless abandon.;)
New kernel! DIY
Lets download this brand new stable mainline kernel - https://www.kernel.org/ (for all options)Code:wget https://www.kernel.org/pub/linux/kernel/v3.x/linux-3.12.tar.xz
..and extract it to /usr/srcCode:sudo tar -xvf linux-3.12.tar.xz -C /usr/src/
..and go thereCode:cd /usr/src/linux-3.12/
lets remove EVERYTHING that is not in Your computer, or rather, build the kernel for this specific HW configuration - guess this is most reasonable for cruncherCode:sudo make localmodconfig
..since my opty has 12c I can put them all working - its available cores/threads +1Code:sudo make -j13 deb-pkg
up one levelCode:cd /usr/src/
lets deploy that freshly baked niceness, [TAB] press the button hereCode:sudo dpkg -i linux-image-3.12.0[TAB]
that tooCode:sudo dpkg -i linux-headers-3.12.0[TAB]
time for restart!Code:sudo shutdown -r now
Now, on boot options screen - select the new kernel from "advanced options".
Profit - if its not fast or has some issues then you can always revert to previous, again on "advanced options" on boot.
[the above is in no way 100% correct way, as im also a fresh convert to linux - but i think/hope any errors will be pointed out]
The above has been tested by me on my cruncher, which is still alive and has gained (with personalized kernel/frequent updates) ~15% (6,5K to 7,5K avg ppd)
If all is good with the new kernel - then lets remove the old one(s)
Gives the list of currently in the systemCode:dpkg --list | grep linux-image
Remove old ones not neededCode:sudo apt-get purge linux-image-[insert old kernel number here]
Update bootmenuCode:sudo update-grub2
Profit!
question: why would I see BoincTasks showing 94% (typical) for each wu when top shows 15*100% and the other 98% in use by wu's
I have watched without boinc running and seen maybe 8-10% of a thread used by other processes.
also, in an ideal world, I would expect to see (again with top) "user" at a v. high percentage and "nice" very low %. I actually see the opposite. Can we alter the cpu priorities within boinc to "normal" and maybe let "nice" do all that work because just now there seems to be a clash between the two systems.
I get ~99,9% for MCM
http://i.imgur.com/ZbPNxlV.jpg
http://i.imgur.com/SrZO3gT.jpg
I get (got) 94.6% on one rig which I just re-booted to have a look at what was going on... meanwhile I have that work suspended and a bunch of high priority work started which is running at 99.46-99.89% :shrug:
Just noticed on another rig a wu at 47% cpu at 99.81% and runtime so far of 13hrs 50mins :eek:
Lots of variance for these WUs.
Seem to run ok though, for me. CPU temps are the same.
There is some thing called "nice level". Looks like you can change that in the cc_config file.Quote:
Originally Posted by OldChap
Edit: forgot to add the link.
http://boinc.berkeley.edu/wiki/Client_configuration
Look for "no_priority_change"
lol... and i've had cron script to do this :rofl:Quote:
Originally Posted by Rob_B
should RTFM before i do things the hard way
"nice" is the priority that BOINC tasks run at, which is why you see so much CPU usage in that category. Threads with a "nice" priority stop to allow other threads to use the CPU, which is how BOINC gets out of the way to allow you to use your system. "User" priority threads are those that you start deliberately, such as your browser, email client, image viewer etc. If you jack the BOINC process priority up to "user" level you will find the system much less responsive to your own tasks and usage and harder to administer.Quote:
Originally Posted by OldChap
Strictly speaking, the whole thing is called "nice" and BOINC worker tasks have a "nice" of 20, meaning they are very nice and get out of the way of everything, while "user" tasks have a lower number, meaning a higher priority, and they are more selfish. Tasks listed as "nice" in TOP etc are just one step higher priority than "idle".
I tried it on one rig and looks like it changes some thing. Only the CEP2 is still on the lower priority. :confused:
http://www.robbakker.com/images/top1.png
Well first of all I had not seen that before so thanks Rob.
D_A: I found the following comment somewhat edifying: Nice is helpful when you are competing against other users on a single machine, but for a single user with multiple jobs, let the scheduler do the scheduling.
I have changed mine now on the dedicated crunchers... just to see what happens. I am not expecting a lot but who knows
EDIT: I can't get screenshot working but I now have 99.8or9% user and nil nice on the one rig I checked so far
http://img.techpowerup.org/131111/1111000000.jpg
They are all Linux. I have noticed many times with BoincTasks that when the numbers are low and I watch it for awhile they come up slowly. I guess it does it's own averaging or some scoccery I do not know. Top is what you should go by. X, the GUI does take some CPU so if you are using the machine moving windows around, opening and closing them it will take some from Boinc tasks.
Installed Mint 64 Cinnamon. I can see why people like it. I was a little apprehensive when I was not given the opportunity to use parted, I may have missed it. Pretty neat trick doing a side by side install. I was pretty aggravated at first as the menu was choppy and jerky but it was the theme I was using, others were fine. Copied /var/lib/boinc-client over from the Ubuntu install after installing from repositories and copying boinc* from the 7.0.65 archive. Ran into ownership problems. Solved them with
while in /var/lib/boinc-client/Quote:
sudo chown -R boinc:boinc *
Picked up where it left off.
BTW - when copying the directories do not worry about errors about the sym links. They were all made when you install from repositories: sudo apt-get install boinc-client. :up:
Yes, you're right. You missed it. You can partition manually if you want, it's the bottom option of the same page where you chose to install side-by-side.
More about "nice"
My linux install on my "Spiders" rig was new and that has seemed to perform very well anyway but the longest wu that I have yet run is just that bit better with nice off and the OS scheduler doing the scheduling. (don't ask me which scheduler is in use here)
http://www.lakecityquietpills.com/ph...9269332101.png
The loss then is 0.07 of an hour in 29 hours = ~4 mins :)
My Arachnid rig is using a linux install that was previously in use on another rig. It was doing OK running in the 97% region. not so great but again changing Nice has now put this into the 99% range.
No pic this time but has lost 58 seconds in 5 hours 13 mins on the current longest wu.
Then we get to the rig with the single processor. This is an HDD that has gone from rig to rig, has been updated, upgraded and generally needs a fresh start it seems. This one continues to run in the 94% range with 16 cores.
switching to 15 cores running pushes the performance back into the 99% range but even though those 15 cores show at 100% and there is no other cpu use seen in top beyond the odd 1-2% used by sys and cinnamon there is currently only 6% showing idle.
This one is going to need work (even if only a fresh install etc)
@OC How did you change the nice value? Can you show a screenshot of top running showing all CPUs? With top running hit 1. PLease. I am experimenting as well using the cc_config.xml way and am not seeing any difference in the %ni or %id columns.
Thanks
---------------------------------------------------------
The top man page does not define all the columns when you hit 1 to see all CPUs.
us: user cpu time (or) % CPU time spent in user space
sy: system cpu time (or) % CPU time spent in kernel space
ni: user nice cpu time (or) % CPU time spent on low priority processes
id: idle cpu time (or) % CPU time spent idle
wa: io wait cpu time (or) % CPU time spent in wait (on disk)
hi: hardware irq (or) % CPU time spent servicing/handling hardware interrupts
si: software irq (or) % CPU time spent servicing/handling software interrupts
st: steal time - - % CPU time in involuntary wait by virtual cpu while hypervisor is servicing another processor (or) % CPU time stolen from a virtual machine
http://img.techpowerup.org/131112/_0000000.jpg
The top one is with <no_priority_change> undefined. The lower one is with <no_priority_change>1</no_priority_change>
The top one is running 2 CEP2 work units, only time I ever really see anything in the %sy column over .ox, or more than less than 1%.
Not seeing criticism but we are possibly talking at crossed purposes.
When I am talking about seeing 94% I am referring to the Boinctasks screen which, oddly enough, I am tending to go with because it seems to accurately fit with the amount of lost time (the difference between real time and cpu time for a wu.)
All the while there is 94% showing in Boinctasks there is 100% showing on all bar one thread (98%) in top.
When I was referring to the idle time, that was seen in the info across the top of top: user, system, nice, idle, wait etc.
All together this is saying to me that if it is not the OS being inefficient, seemingly the case judging by the 100% for each thread showing but there is only 6% of idle time when running 15 threads then the losses must lay elsewhere and perhaps a new install would rule out less than optimum "drivers" etc.
It could be lesser hardware causing this but....
I am just following my nose on this and my logic may be faulty too bearing in mind my lack of detailed knowledge on how this stuff actually works..... but it should only cost an hour to find out.
EDIT: I have an issue trying to get a screenshot... what I get just now is a screen that is older than the current screen. any tips on that and I will gladly post.
I changed my value using boinktasks> extra> edit config file> <no_priority_change>1</no_priority_change>
I tried just getting it to read the config file again but had to actually re-start boinc for it to work
You could do this with gedit and just add the above line to cc_config. (listen to me talking to the bearded one ....) :rofl:
I was wondering if any of you Linux guys who have switched over to the newest BOINC client for 64 bit Linux (7.2.28?) have seen a difference in crunching times as compared to 7.0.65 or .64?
Also, as a Linux newb, I am still learning things about the OS. To that end, did you or do you have to use a command line way of updating to the new version or did you have to uninstall the current BOINC client and then install the new one?
sorry OC I changed my post above. The more I got to thinking about it all I thought I did not really understand things as well as I thought. Still not sure. Why are all processes showing %id when they show no reason for it. no IO wait states, no system or %sy, except for CEP2. I used to have to deal with this stuff on very busy database servers with hundreds of users. Not sure it this is all different, Linux is that different than Solaris in how it reports these things, I am old and with foggy brain and beginnings of dementia. Maybe some of all three.
Will be interesting to see what we learn. :up:
I have seen no difference but the newer one allows you to use app_config.xmlQuote:
Originally Posted by yojimbo197
see this thread on how to update and or install newest Boinc.
Jim, Command line is the easiest way I know Poppa posted at start of his update linux thread or there is the D_A post on the using Linux thread
http://www.xtremesystems.org/forums/...ely-site-Linux
http://www.xtremesystems.org/forums/...=1#post5024891
whichever you like.
hard to tell if any improvement due to change to MCM1
Poppa, edit to my previous too
I'd have to look it up but all the reporting stuff in Solaris is available in the Linux version of top, you just need to use the right flags ... which I don't remember. I'm sure you can configure top's default to show whatever you want, but again I don't remember the syntax etc. man top is your friend.Quote:
Originally Posted by PoppaGeek
When you run top as above do you see about the same thing?
I'll steal rob's ss for this:
http://www.lakecityquietpills.com/ph...4658855105.png
on mine 1. has gone to 90+%
2 and all in column 3= 0%
4. all 100%
If you have added the line to cc_config then quit and re-start Boinc and yours should change too
installed shutter
http://lakecityquietpills.com/photo/...1216932189.png
I think that all commands can use -h for a help syntax screen