CentOS 5 :: Openmpi Cannot Run On 5.6
Apr 19, 2011
I find problem when run MPICH2 (also Openmpi) on the newest issue of centos-5.6 prerevious I tried MPICH2 on Centos 4, there is no such problem. My procedure is as follows: I built a Bewolf Cluster between 6 nodes of our serve (HP
Proliant G460 G7), and installed the MPICH2 on one node (assuming at /mirror), ./configure --prefix=/mirror/mpich CC=icc CXX=icpc F77=ifort FC=ifort make 2>&1 make install 2>&1
using NFS, the directory of /mirror was successfully exported to the rest of 5 nodes. Now as I test the mpirun, it runs very well on a single node, however it hangs across multiple nodes. Also I have built the passphraseless ssh-keys connection between each other. Now the problem is that even I use the mpd, boot the nodes(those nodes are in the list, and there is no warning during the booting), however when I run the mpirun, whatever how much n (-n X)I set, all the process will be run at the local nodes. I am not sure what are blocking the data exchange between the nodes, by the way, I have set the iptables to accept then data.
View 1 Replies
ADVERTISEMENT
Oct 18, 2010
how can i install openmpi.taz.bz2 on debian linux?
View 3 Replies
View Related
Jan 13, 2011
I am taking a parallel computing course at school and will be required to write programs using MPI soon. All of the school servers have Lam/MPI installed on them but I can only find OpenMPI slackbuild scripts. If I install OpenMPI on my computer and do my development on my own machine, will I have much trouble when it's compiled/run on the school servers?
View 5 Replies
View Related
Mar 9, 2009
While installing openmpi it is requried to mount using following command:mount nfs yourmachineIP /mnt/nfs but executing this command gives following error: mount: directory to mount not in host:dir format although i m able to mount a folder using mount �t nfs yourmachineIP:/dir /mnt/nfs where /dir is specified in /etc/exports. Is it possible 2 mount only using IP? in that case what should be mentioned in /etc/exports?
View 1 Replies
View Related
Jul 24, 2010
I'm trying to get OpenMPI (a parallel programming library) working on my home system. I have just two machines on it now, t61 and quad, connected through a router. (Which is also connected to cable modem for internet.) I can ssh between the machnes, mount directories with NFS, etc. However, I just can't get the OpenMPI to run. The OpenMPI message board suggested that the most probable cause is that the firewall is blocking TCP. I don't know how to tell if that's the problem, and can't find any manual for the SuSE firewall, while the various Wikis &c that pop up in a search don't provide any information that addresses my problem.
View 9 Replies
View Related
Mar 26, 2011
I have installed OpenMPI from ubuntu repository. However some users need openMp. The problem is I didn't find that in ubuntu repository. Where can I found deb packages? If I must install from source, then is there any conflict between openMp and openMPI?
View 4 Replies
View Related
Feb 4, 2010
I've installed openMPI-1.4.1 on my laptop and I'm trying to compile some fortran code I've written using MPI. I downloaded the openMPI-1.4.1 file from the openMPI website and used the shell script they provide to install it. This is the shell script:
shell$ cd openmpi-1.4.1
shell$ ./configure --prefix=/usr/local
shell$ make all install
There was an additional line extracting the tar file, but I did this by hand. The first time I ran this script there was an error on the final stage saying that a directory couldn't be created because of lack of permission. I decided to run the script again this time as a sudo.
Having run the script again everything appeared to install properly, however when I went to compile some fairly simple code I got a message saying that the compiler couldn't find MPI. I was using this command: gfortran messages.f90 -o messages I looked on the openMPI website where it recommends compiling with a compiler called mpif. However when I type mpif messages.f90 -o messages I get an error message saying mpif isn't a recognized command......
View 2 Replies
View Related
Aug 4, 2010
Problems with xen 3.1.2 creating and installing domU CentOS 5.5 I am trying to do a virt-install --prompt --paravirt, and the install error out while trying to download install RPMs. I have googled for xen centos virt-install and various permutations. I have found similar howtos for 5.4 and 5.3, but they resulted in the same outcome. I did find several posts with similar issues, but no solution was posted.[URL]..
View 4 Replies
View Related
Feb 9, 2011
I am in the process of building a new server on an Asus P5QPL-AM motherboard and an Intel E8600 processor.explain to me the difference between the two versions and what would you recommend.Also, is there any advantage of SATA over IDE hard drives?
View 1 Replies
View Related
Jan 1, 2010
I have two servers, both running CentOS, but one running Cpanel as the control panel and the other running Webmin.
I would like to sync the web folders using rsync. However, when I sync the files from the Cpanel host the permissions look as follows:
Code:
But that doesn't work on the Webmin server which needs the following:
Code:
It seems the Cpanel install has the group as 'nobody' whilst webmin has it the same as the file owner. I can fix this by running:
Code:
But as the file transfers will be frequent, I don't want the files to be inaccessible until chown is run.
So, what I really want to know is - how can I change things on the 'webmin' server so that files with ownership owner:nobody will run?
View 1 Replies
View Related
Feb 15, 2009
The installer can't see my raid controller (I assume) as I'm getting the following error:"Error opening /dev/mapper/isw_jbhgjgjj_Vol0: No such device or address"It just sees them as 4 individual drives: sda, sdb, sdc and sdd.Please note that I have set up the RAID 5 in the controller bios interface and the image name is Vol0, which it seems that it tries to load but for some particular reason it can't.I have also tried different bios settings and nothing worked.
View 3 Replies
View Related
Apr 28, 2009
I'm running a fresh install on an intel i7 system on an asus p6t deluxe v2 motherboard with the onboard NIC (Marvell Technology pci-e). I know the NIC is working as when I boot into the onboard OS that Asus provides (Asus Express Gate SSD)I have a working net connection.Booting up CentOS hangs at determining IP information for eth0 and eventually fails with the following error:PING xxx.170.30.1 from xxx.170.31.231 eth0: 56(84) bytes of data---xxx.170.30.1 ping statistics ---4 packets transmitted, o received, +3 errors, 100% packet loss, time 2999ms, pipe 3 failed.
RTNETLINK answers: file exists.
RTNETLINK answers: file exists.
RTNETLINK answers: file exists.
[code]....
View 14 Replies
View Related
May 5, 2009
I'm trying to install a dual booting machine with OpenSUSE v11.1 32bit and CentOS v5.2 64bit. I installed OpenSUSE first and allowed it to install and configure grub in the MBR and after that I wanted to proceed with CentOS v5.2. The installation went fine with two notable exceptions:- when I had to configure grub installation parameters, CentOS offered me only 2 solutions: either install it on the MBR of the first hard disk or not installing it at all. Other distributions are more flexible allowing you to install it in the boot sector of the root partition for example. Because I didn't want to ruin the existent grub onfiguration, I reluctantly accepted not to install it for CentOS assuming that I could manually configure the entry later in grub's menu.lst file.
- when I was presented with the options for software components installation, I've clicked on virtualization category/function because I intend to use the machine as a VMware host. There was no guidance on screen at that point and I blindly assumed that by choosing the virtualization function I would get necessary tools and drivers that will help me further on. It seems that this was a wrong move as you can see it below.
After completing the installation, I tried to search for a template or guiding on how the menu entry in menu.lst should look like but the grub directory was empty, not surprisingly because I've told CentOS earlier not to install it. Using the files in the /boot directory from the CentOS installation I tried to improvise a menu entry but it's not working. The boot stops with famous Error 13: Invalid or unsupported executable format. Using the file command to check what kind of files I'm trying to load as kernels I'm getting :
marte:~ # file /mnt/vmlinuz-2.6.18-92.el5xen
/mnt/vmlinuz-2.6.18-92.el5xen: gzip compressed data, from Unix, last modified: Tue Jun 10 19:20:51 2008, max compression
[code]....
View 2 Replies
View Related
Jul 9, 2009
The first server I installed installed fine. The second server, installed with the same config, went to "kernel panic not syncing no init found try passing init= option in kernel" error. I tried reinstalling but it keeps going to that error after install reboot. The storage is ISCSI connected via Intel Server Adapter, which allows it to boot from ISCSI. Not sure if that's the cause for the problem, but the first server is connected to the same ISCSI and installed just fine.
Is there a way that I can make sure ISCSI module installs during installation? Although I think it is installed since it's able to copy the files and setup /dev/sda. I just wana make sure that it installs during setup.
View 1 Replies
View Related
Aug 14, 2009
I'm trying to create a CentOS based liveCD that mounts a NFS share and executes one script there. The NFS never mounts altough while logged I can mount it with the very same command that I use in the KS.
In the %post I have :
%post --log=/tmp/post.log --erroronfail
mkdir /mnt/nfs
mount -o nolock 10.23.1.1:/csc/RemoteHome /mnt/nfs
I added the --log in order to debug, but nothing is written in /tmp/post.log. I tried redirection on the mount command with >> /tmp/debug.log but this is not written. Maybe during the post sequence /tmp is RO ? Anyway, I tried with ifup lo, service portmap start in the %post, but doesnt change (I even have a service : command not found in the live cd creator output).
Here is the remainder of the KS :
lang en_US.UTF-8
keyboard us
timezone US/Eastern
[Code]...
View 10 Replies
View Related
Oct 30, 2009
I've got two USB webcams I'm trying to get running with CentOS 5.4 x86_64. I've downloaded and installed the gspca driver from http://mxhaard.free.fr/spca5xx.html. Running lsusb yields:
Bus 002 Device 003: ID 0ac8:307b Z-Star Microelectronics Corp. USB 1.1 Webcam
Bus 002 Device 002: ID 0ac8:307b Z-Star Microelectronics Corp. USB 1.1 Webcam
View 9 Replies
View Related
Dec 21, 2009
I have installed CentOS 5.3 on my PC which has an Atheros AR8121/8113/8114 Gigabit Ethernet. Unfortunately 5.3 has no drivers for this LAN Card. I searched the Google for drivers but was unable to find any. Does the 5.4 has them? If yes then how can i upgrade it to 5.4 without Network connectivity?
View 11 Replies
View Related
Feb 28, 2010
I built a home server (NAS/WWW/SSH/media server etc) and chose CentOS 5 as the OS (stability, easy of configuration).I was just about to start tuning the power consumption when I realised that the kernel CentOS uses is so "old" that it does not support the latest reduced power consumption enhancements that Linux has achieved in big strides in the recent past (we are probably still talking 6-12+ months ago e.g. tickless kernel)..
So my questions; 1) I know CentOS was maybe not meant for home servers (certainly its not its primary purpose), but if it is, any ideas of what kind of power consumption it takes (I know its relative) and if there are particular power consumptions that are worthwhile?
2) Do you recommend me compiling my own 2.6.21+ kernel from kernel.org or am I just likely to have compatibility issues (I really did not want to do that) or when is CentOS 5.4 supposed to have a newer 2.6.21+ version kernel?
Was it wrong of me in principle to choose CentOS for a home server when I am power conscious? (I don't have a low-power VIA processor either but a P4 so I am really just hoping to make do with software changes).
View 4 Replies
View Related
Oct 12, 2010
I have a few mail servers, a mail log server and a web server running on Centos 5. Now I have a task: to avoid accidental crashes on the production servers while installing updates, my boss asked me to do clones (these clones will all be VMware virtual machines) of the servers (EXCLUDING the actual e-mails and log contents) and then to run those clones on VMWare Server. This way, first I will install and test updates on the clones and - if they will be running without crashes - I will apply the updates on the real production servers themselves.
I have already installed VMWare Server 2.0 I have a few questions: How do I build the virtual machines to exclude the actual mail files and mail logs? Can I use VMware Converter for this purpose, or do I have to use another program? How do I actually do this cloning? Is there a tutorial on how to do this?
View 4 Replies
View Related
Jan 27, 2009
I have a Centos 5.2 VM running under Windows 2008 Hyper-V. If I add two "legacy Nic's" and bridge them, everything appears to work fine. However, if I remove the legacy nics and replace them with two synthetic nics (linux integration components), the nics seems to function fine UNTIL I add them to a bridge.This VM will eventually be used as a DansGuardian/Squid transparent proxy for a local library. The logical layout is represented by the attached jpg image. I have also included the ifcfg- scripts for the adapters and bridge for both scenarios.It would also be useful if someone else has Hyper-V with the integration components and dual physical nics to see if they can reproduce my symptoms on their box.Legacy Adapters:
ifcfg-br0:
DEVICE=br0
BOOTPROTO=dhcp
[code]....
View 2 Replies
View Related
Mar 17, 2009
If I ssh from my laptop (running F10) to the server (centos 5.2) it asks for the password, but everytime I enter the correct password it says incorrect password. when I do the same from the server to my laptop I can get in just fine. I think my passwords are stored as ssha in the LDAP (I tried clear passwords and that dosen't work either).
View 1 Replies
View Related
May 1, 2009
I just have CentOS installed and cannot get the Agere et131x network card to work.I have tried the following:Reboot the computer and still did not see the network hardware.
Fai
lspci
Quote:
[code]....
View 7 Replies
View Related
May 3, 2009
I have tried to install debian 5.0 and 4.0 without any luck. I have both tried virt-manager and virt-install with both debian-501-i386-netinst.iso, and full dvd. I't always gives me the same error:Starting install.Could not find an installable distribution at '/xen/debian-501-i386-netinst.iso'
[Code]...
View 2 Replies
View Related
May 4, 2009
I was try to compile the drirver from cd asus, but my lan card still not detected in centos 5.3. when i bringing up eth0 its show like : "the atl2 device eth0 does not to be present, delaying initialization."iwas suft on google some driver but all driver just to centos 5.3 when i install the rpm it does not work. i love centos i dont want to change my lovely centos.
View 10 Replies
View Related
Jun 8, 2009
i didnt find a solution to a display problem im having. my server mobo is an asus k8n-dre with built in video. after the install(using built in video) the screen dosnt display fully as in black sections on the left and top of the screen. from the searching i did i found something about an "xrandr" command that would let me change the video settings but it didnt do anything with the black areas when i tried it. i also have a nvidia geforce 9500gt video card so i changed the jumper to boot from it and tried again. the black sections are still there but the centos display covers much more of the screen now. theres still about a third of an inch gap on the left hand side and about 1/4 of an inch on the top. ive also tried updating the software to see if it may be a driver issue but no luck. anyone know how or if it can fit to full sceen? the screen displays up to 1440x900 if im not mistaken but with the built in video it gave me lots of display options including the 1440x900. after changing to the nvidia card it only shows 800x600 and 640x480.
View 5 Replies
View Related
Jun 24, 2009
I try to set up a 5 nodes cluster and a shared Coraid Storage with conga but it fails with "Shared Storage Support" checked.The message is:'A problem occurred when installing packages: Packages of set "Clustered Storage" are not present in available repository' and it is shown under every node on the next sceen after I submit.The pc where conga runs is on the same subnet (192.168.xxx.xxx) and it has the same /etc/hosts of the other nodes.In that pc runs a proxy too and the nodes go out through it (that pc has 2 NICs)Every node (2.6.18-128.1.14.el5-xen-x86_64) is patched whith the last yum update (this morning) the same is for the pc (2.6.18-128.1.14.el5).Every node has 4 NICs , 2 NICc towards the storage the others in bonding towards the WAN.Every node is exactly alike, they have been installed with the ks.cfg generated from the first node and they all have the support for Clustering, Virtualization, and Clustered Storage.
View 1 Replies
View Related
Jun 27, 2009
This will be a little long (having read Phil's 'how to ask questions' FAQ). I'm trying to get OpenVPN working between my CentOS server and some Windoze laptops running XP. There seems to be plenty of sample
config files available, but to date, none of them have worked for me. Pulling out my trusty Wireshark, I've found some clues,
BACKGROUND:
My local subnet (NAT'ed by my gateway router) is 192.168.52.x. My router has been configured with a conduit (port-forward) for port 1194 (the standard OpenVPN port), which points towards my CentOS server.The CentOS server is .52.112, and the supplicant is .52.110. I have tried the lient both inside and outside my local subnet, with no difference in events or outcomes.
[Code]...
View 2 Replies
View Related
Jul 2, 2009
Why is cx88-dvb no longer included in the CentOS 5 kernels found in centosplus?
It is present in all 2.6.18-92.1.x versions, but not in 2.6.18-128.1.x versions.
View 3 Replies
View Related
Jul 8, 2009
I've installed CentOS 5.3 on a machine, and I need a Samba version 3.2 or higher. Since 3.4 is out, I thought I'd grab that. But, "yum list|grep samba" gives me only version 3.0.33. Is there a package of Samba I can grab that will upgrade the 3.0 installation so that I don't have two laying around? If not and I need to compile from source, do you have any suggestions for what arguments I should give configure? I'm not used to Linux coming from the BSD world
View 1 Replies
View Related
Oct 3, 2009
i find this is an issue which might be tiresome for everyone except for those experts. Esp. for the computers which are dual booting(both windows xp and centos installed in the same computer), that when we are in CentOS we need to switch onto windows immediately. and vice versa. i see this is possible only by shutting down the centos/windows and then restarting onto the other OS.
Can anyone tell me the the solution to work simultaneously both xp and centos without the need of shutting down and restarting There maybe or may not be the solution...but i find the pleasure to know if there is for example, remote desktop connection in windows xp, is there any solution, we can work on windows via centos or vice versa ( in the same machine),
View 2 Replies
View Related
