October 2013 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] mpich2 installation fails
by bhimashankar dhappadhule 24 Oct '13

24 Oct '13

sir,i installed mpich2 on ubuntu 12.04 on three machine of my college.i need to form cluster please provide me steps to from cluster. sir its urgent, thank you. On Wed, Oct 23, 2013 at 12:09 AM, Ken Raffenetti <raffenet(a)mcs.anl.gov>wrote: > This output is incomplete. Were you sure to pipe all output from configure > into a separate file? e.g. > > ./configure | tee c.txt > > The config.log file that is generated in your MPICH directory during > configure would also be helpful. > > ----- Original Message ----- > > From: "bhimashankar dhappadhule" <dhappabn(a)gmail.com> > > To: discuss(a)mpich.org > > Sent: Tuesday, October 22, 2013 11:55:24 AM > > Subject: Re: [mpich-discuss] mpich2 installation fails > > > > > > > > sir,i am sending my configuration file.plz help us urgently. > > > > > > > > On Tue, Oct 22, 2013 at 7:43 PM, Ken Raffenetti < > > raffenet(a)mcs.anl.gov > wrote: > > > > > > Hi, > > > > Can you send us the output of your configure command, as well as the > > config.log file that was generated? Additionally, if you do not need > > the Fortran bindings for MPICH, you can try adding --disable-f77 and > > --disable-fc to your configure and see if your build completes then. > > > > Ken > > > > > > > > ----- Original Message ----- > > > From: "bhimashankar dhappadhule" < dhappabn(a)gmail.com > > > > To: discuss(a)mpich.org > > > Sent: Tuesday, October 22, 2013 6:59:06 AM > > > Subject: [mpich-discuss] mpich2 installation fails > > > > > > > > > > > > > > > > > > sir , > > > while installing mpich2 on ubuntu 13.04 ,configuration successfully > > > done but at time of command '' make ",i get error like shown in > > > attachment.I urgently required your help sir. > > > Thank you...! waiting for reply. > > > > > > _______________________________________________ > > > discuss mailing list discuss(a)mpich.org > > > To manage subscription options or unsubscribe: > > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Re: [mpich-discuss] machinefile question and select hydra as primary pm
by Wesley Bland 24 Oct '13

24 Oct '13

Hi Mark, On Oct 24, 2013, at 3:07 AM, mark <dimitsas.markos(a)gmail.com> wrote: > Hello, > i did a fresh install of 3.0.4. on a linux laptop and since i dont have much experience i followed the installation readme on the tar file and everything went well expect the part when i test the installation. The first test run : mpiexec -n <number> ./examples/cpi , went as planned but then the second test run on multiple nodes (mpiexec -f machinefile -n <number> ./examples/cpi) returned some errors. As the readme states there should be a machinefile. The thing is i cant find this file and tried to create it my self in the given form, just using only my laptops hostname a couple of times and placing it in the mpich installation directory ( /home/<username>/mpich-install) and that didnt fix the errors. A machine file is necessary if you have multiple machines with MPICH installed and you want to run an MPI application across all of them. If you don’t have it installed on each machine, it won’t work. The normal use case is to have a cluster with an NFS shared directory where the installation resides (such as your home directory). On a network of laptops, this probably won’t be the case so you’ll have to install things manually. Note that if you’re using Linux, chances are that you can just install MPICH via your package manager so you don’t have to worry about prefixes, paths, etc. The machine file just tells your local machine (where you’re starting the MPI application) where to try to execute the remote processes. It is just of the form: host1 host2 host3 <etc.> And is a flat file wherever you want to put it. To use it, you use the flag -machinefile <path_to_machinefile>. > > One other thing i noticed is that on the readme states that hydra is the primary process manager, but using the mpiexec -info it clearly states that pmi ( or something similar ) is the primary pm. To be honest, it probably makes very little difference to you what process manager you’re using unless you have a very specific need. All of that should be transparent to you. > > And for a last thing my installation was made without using any prefixes at the configuration. Again, if you use a pre-built package, this will solve the problem, but otherwise you should either reinstall with a prefix or go add the directory where MPICH was installed by default to your PATH and LD_LIBRARY_PATH. I believe it’s /usr/local/bin and /usr/local/lib respectively. You don’t need to install as root to make this work. > > PS.One other thing i just noticed is that when i did the installation i could run mpich ( the test runs with the pi programm, mpiexec -info , etc) , but now after a reboot i cant run mpich on the terminal. For example mpiexec -info returns bash: mpiexec: command not found . Whats the deal with that? Didnt i installed mpich on my machine? Same problem as above. > > Thank you for your time. Thanks, Wesley > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH2 on LAN with Ubuntu 12.04
by heshsham basit 24 Oct '13

24 Oct '13

I get this on both the machines: root@ubuntu:/home/# getent hosts ubuntu 192.168.1.4 ubuntu root@ubuntu:/home/# getent hosts hesh 192.168.1.5 hesh root@ubuntu:/home/# which is expected- am I right? (one of my IP address changed as I connected to a different WLAN) On 24 October 2013 09:02, Reuti <reuti(a)staff.uni-marburg.de> wrote: > Hi, > > Am 24.10.2013 um 12:05 schrieb heshsham basit: > > > Thanks a lot for your replies. > > I followed the instructions given here > > wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager1 > > > > for running MPI program on multiple machines: > > > > My master node is : 192.168.1.4 > > My client is: 192.168.1.6 > > You defined them both on both machines in /etc/hosts? What should give a > valid reply is: > > $ getent hosts ubuntu > 192.168.1.4 ubuntu > $ getent hosts hesh > 192.168.1.6 hesh > > resp. the on the other machine. > > -- Reuti > > > > I gave the following command from 192.168.1.4 > > > > root@ubuntu:/home# mpiexec -f hosts.cfg -n 4 hello > > root(a)192.168.1.6's password: > > [proxy:0:0@hesh] HYDU_sock_connect (./utils/sock/sock.c:171): unable to > get host address for ubuntu (1) > > [proxy:0:0@hesh] main (./pm/pmiserv/pmip.c:209): unable to connect to > server ubuntu at port 42104 (check for firewalls!) > > > > Why I am getting there errors? > > > > Both the machines have locally generated executable hello. Do I must > have same executable copied in both machines? > > > > On master node 192.168.1.4 > > root@ubuntu:/home# ls > > acme hello hosts.cfg mpi_hello.c > > > > On client 192.168.1.6 > > > > root@hesh:/home#ls > > hesh hello > > > > hello in each machine sis generated after compilation locally. > > > > > > > > On 22 October 2013 12:51, Jeff Hammond <jeff.science(a)gmail.com> wrote: > > See > http://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager > > for how to launch MPI across multiple machines. You do not need to > > use any special hardware adapter; any Ethernet card should provide > > sufficient TCP/IP support. > > > > Jeff > > > > On Tue, Oct 22, 2013 at 9:23 AM, Pavan Balaji <balaji(a)mcs.anl.gov> > wrote: > > > Hello, > > > > > > You'll need to either install mpich on a shared file system (such as > NFS) or individually on each of the nodes. The README file in the mpich > tarball should have most of this information. > > > > > > -- Pavan > > > > > > On Oct 22, 2013, at 8:26 AM, heshsham basit wrote: > > > > > >> Hi, > > >> > > >> I did some experimentation with MPICH on windows, and recently came > to know that there is no support for Windows now. > > >> I am now shifting to Ubuntu. I was successful in installing MPICH2 > and running my first MPI program on a single machine. > > >> > > >> Now I want to run my code on multiple PCs. How do I install MPI in a > LAN ? What are the additional requirement if any? Do I need any specific > hardware -network adapter My card details are as under: > > >> > > >> root@ubuntu:/home# lspci -nnk | grep -i net -A2 > > >> 01:07.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. > RTL-8139/8139C/8139C+ [10ec:8139] (rev 10) > > >> Subsystem: Hewlett-Packard Company Device [103c:301d] > > >> Kernel driver in use: 8139too > > >> > > >> > > >> Most of my machines are having this type of card and are connected in > LAN. > > >> Really appreciate for any pointers to appropriate document or > suggestions. > > >> > > >> > > >> Thanks and regards, > > >> > > >> > > >> > > >> > > >> _______________________________________________ > > >> discuss mailing list discuss(a)mpich.org > > >> To manage subscription options or unsubscribe: > > >> https://lists.mpich.org/mailman/listinfo/discuss > > > > > > -- > > > Pavan Balaji > > > http://www.mcs.anl.gov/~balaji > > > > > > _______________________________________________ > > > discuss mailing list discuss(a)mpich.org > > > To manage subscription options or unsubscribe: > > > https://lists.mpich.org/mailman/listinfo/discuss > > > > > > > > -- > > Jeff Hammond > > jeff.science(a)gmail.com > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Re: [mpich-discuss] machinefile question and select hydra as primary pm
by Reuti 24 Oct '13

24 Oct '13

Hi, Am 24.10.2013 um 14:07 schrieb mark: > Στις 24/10/2013 02:47 μμ, ο/η Reuti έγραψε: >> Hi, >> >> Am 24.10.2013 um 10:07 schrieb mark: >> >> >>> i did a fresh install of 3.0.4. on a linux laptop and since i dont have much experience i followed the installation readme on the tar file and everything went well >>> >> >> You mean a plain `./configure && make && make install`, hence installed at the standard location as root? So you used --prefix=... - fine. >> Well i mean that actually , but as non-root. And in the user folder as in /home/<username>/mpich-install . >>> expect the part when i test the installation. The first test run : mpiexec -n <number> ./examples/cpi , went as planned but then the second test run on multiple nodes (mpiexec -f machinefile -n <number> ./examples/cpi) returned some errors. As the readme states there should be a machinefile. The thing is i cant find this file and tried to create it my self in the given form, just using only my laptops hostname a couple of times and placing it in the mpich installation directory ( /home/<username>/mpich-install) and that didnt fix the errors. >>> >> >> Is MPICH 3.0.4 also available on both machines? >> >> >> > For now i have only this machine.Single laptop and i would like to run some mpi programs. Maybe in this case , perhaps is the machinefile useless?? In case you have only one machine you don't need it necessarily. >>> One other thing i noticed is that on the readme states that hydra is the primary process manager, but using the mpiexec -info it clearly states that pmi ( or something similar ) is the primary pm. >>> >> http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_What_ar… >> >> >> >> Maybe it should better read: "Process Manager uses: pmi", and as the default is to compile Hydra, you know it already (there are different `mpiexec`s for the different process managers and e.g. mpiexec.gforker has no "-info" option). >> >> >> > So the primary processor manager (default) in my installation is hydra right? >>> And for a last thing my installation was made without using any prefixes at the configuration. >>> >>> PS.One other thing i just noticed is that when i did the installation i could run mpich ( the test runs with the pi programm, mpiexec -info , etc) , but now after a reboot i cant run mpich on the terminal. For example mpiexec -info returns bash: mpiexec: command not found . Whats the deal with that? Didnt i installed mpich on my machine? >>> >> Might be a $PATH problem, in case it wasn't installed at the default location. > I thought of that too, and i checked it but unsuccessfully. Maybe it has to do with the fact that i didn't installed it as root.Maybe i need to redo the installation as root? Any suggestions? As to the install location etc ? You need to add something to your $PATH only: $ export PATH=$HOME/mpich-install/bin:$PATH and try again. -- Reuti

1 0

Re: [mpich-discuss] MPICH2 on LAN with Ubuntu 12.04
by Reuti 24 Oct '13

24 Oct '13

Hi, Am 24.10.2013 um 12:05 schrieb heshsham basit: > Thanks a lot for your replies. > I followed the instructions given here > wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager1 > > for running MPI program on multiple machines: > > My master node is : 192.168.1.4 > My client is: 192.168.1.6 You defined them both on both machines in /etc/hosts? What should give a valid reply is: $ getent hosts ubuntu 192.168.1.4 ubuntu $ getent hosts hesh 192.168.1.6 hesh resp. the on the other machine. -- Reuti > I gave the following command from 192.168.1.4 > > root@ubuntu:/home# mpiexec -f hosts.cfg -n 4 hello > root(a)192.168.1.6's password: > [proxy:0:0@hesh] HYDU_sock_connect (./utils/sock/sock.c:171): unable to get host address for ubuntu (1) > [proxy:0:0@hesh] main (./pm/pmiserv/pmip.c:209): unable to connect to server ubuntu at port 42104 (check for firewalls!) > > Why I am getting there errors? > > Both the machines have locally generated executable hello. Do I must have same executable copied in both machines? > > On master node 192.168.1.4 > root@ubuntu:/home# ls > acme hello hosts.cfg mpi_hello.c > > On client 192.168.1.6 > > root@hesh:/home#ls > hesh hello > > hello in each machine sis generated after compilation locally. > > > > On 22 October 2013 12:51, Jeff Hammond <jeff.science(a)gmail.com> wrote: > See http://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager > for how to launch MPI across multiple machines. You do not need to > use any special hardware adapter; any Ethernet card should provide > sufficient TCP/IP support. > > Jeff > > On Tue, Oct 22, 2013 at 9:23 AM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > Hello, > > > > You'll need to either install mpich on a shared file system (such as NFS) or individually on each of the nodes. The README file in the mpich tarball should have most of this information. > > > > -- Pavan > > > > On Oct 22, 2013, at 8:26 AM, heshsham basit wrote: > > > >> Hi, > >> > >> I did some experimentation with MPICH on windows, and recently came to know that there is no support for Windows now. > >> I am now shifting to Ubuntu. I was successful in installing MPICH2 and running my first MPI program on a single machine. > >> > >> Now I want to run my code on multiple PCs. How do I install MPI in a LAN ? What are the additional requirement if any? Do I need any specific hardware -network adapter My card details are as under: > >> > >> root@ubuntu:/home# lspci -nnk | grep -i net -A2 > >> 01:07.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ [10ec:8139] (rev 10) > >> Subsystem: Hewlett-Packard Company Device [103c:301d] > >> Kernel driver in use: 8139too > >> > >> > >> Most of my machines are having this type of card and are connected in LAN. > >> Really appreciate for any pointers to appropriate document or suggestions. > >> > >> > >> Thanks and regards, > >> > >> > >> > >> > >> _______________________________________________ > >> discuss mailing list discuss(a)mpich.org > >> To manage subscription options or unsubscribe: > >> https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] machinefile question and select hydra as primary pm
by mark 24 Oct '13

24 Oct '13

???? 24/10/2013 02:47 ??, ?/? Reuti ??????: > Hi, > > Am 24.10.2013 um 10:07 schrieb mark: > >> i did a fresh install of 3.0.4. on a linux laptop and since i dont have much experience i followed the installation readme on the tar file and everything went well > / > You mean a plain `./configure && make && make install`, hence installed at the standard location as root?/ > Well i mean that actually , but as non-root. And in the user folder as in /home/<username>/mpich-install . >> expect the part when i test the installation. The first test run : mpiexec -n <number> ./examples/cpi , went as planned but then the second test run on multiple nodes (mpiexec -f machinefile -n <number> ./examples/cpi) returned some errors. As the readme states there should be a machinefile. The thing is i cant find this file and tried to create it my self in the given form, just using only my laptops hostname a couple of times and placing it in the mpich installation directory ( /home/<username>/mpich-install) and that didnt fix the errors. > / > Is MPICH 3.0.4 also available on both machines?/ > For now i have only this machine.Single laptop and i would like to run some mpi programs. Maybe in this case , perhaps is the machinefile useless?? >> One other thing i noticed is that on the readme states that hydra is the primary process manager, but using the mpiexec -info it clearly states that pmi ( or something similar ) is the primary pm. > http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_What_ar… > > /Maybe it should better read: "Process Manager uses: pmi", and as the default is to compile Hydra, you know it already (there are different `mpiexec`s for the different process managers and e.g. mpiexec.gforker has no "-info" option)./ > So the primary processor manager (default) in my installation is hydra right? >> And for a last thing my installation was made without using any prefixes at the configuration. >> >> PS.One other thing i just noticed is that when i did the installation i could run mpich ( the test runs with the pi programm, mpiexec -info , etc) , but now after a reboot i cant run mpich on the terminal. For example mpiexec -info returns bash: mpiexec: command not found . Whats the deal with that? Didnt i installed mpich on my machine? > /Might be a $PATH problem, in case it wasn't installed at the default location./ I thought of that too, and i checked it but unsuccessfully. Maybe it has to do with the fact that i didn't installed it as root.Maybe i need to redo the installation as root? Any suggestions? As to the install location etc ? > > -- Reuti Thank you for your time.

1 0

Re: [mpich-discuss] machinefile question and select hydra as primary pm
by Reuti 24 Oct '13

24 Oct '13

Hi, Am 24.10.2013 um 10:07 schrieb mark: > i did a fresh install of 3.0.4. on a linux laptop and since i dont have much experience i followed the installation readme on the tar file and everything went well You mean a plain `./configure && make && make install`, hence installed at the standard location as root? > expect the part when i test the installation. The first test run : mpiexec -n <number> ./examples/cpi , went as planned but then the second test run on multiple nodes (mpiexec -f machinefile -n <number> ./examples/cpi) returned some errors. As the readme states there should be a machinefile. The thing is i cant find this file and tried to create it my self in the given form, just using only my laptops hostname a couple of times and placing it in the mpich installation directory ( /home/<username>/mpich-install) and that didnt fix the errors. Is MPICH 3.0.4 also available on both machines? > One other thing i noticed is that on the readme states that hydra is the primary process manager, but using the mpiexec -info it clearly states that pmi ( or something similar ) is the primary pm. http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_What_ar… Maybe it should better read: "Process Manager uses: pmi", and as the default is to compile Hydra, you know it already (there are different `mpiexec`s for the different process managers and e.g. mpiexec.gforker has no "-info" option). > And for a last thing my installation was made without using any prefixes at the configuration. > > PS.One other thing i just noticed is that when i did the installation i could run mpich ( the test runs with the pi programm, mpiexec -info , etc) , but now after a reboot i cant run mpich on the terminal. For example mpiexec -info returns bash: mpiexec: command not found . Whats the deal with that? Didnt i installed mpich on my machine? Might be a $PATH problem, in case it wasn't installed at the default location. -- Reuti > Thank you for your time. > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH2 on LAN with Ubuntu 12.04
by heshsham basit 24 Oct '13

24 Oct '13

Hi Jeff, Thanks a lot for your replies. I followed the instructions given here wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager1 for running MPI program on multiple machines: My master node is : 192.168.1.4 My client is: 192.168.1.6 I gave the following command from 192.168.1.4 root@ubuntu:/home# mpiexec -f hosts.cfg -n 4 hello root(a)192.168.1.6's password: [proxy:0:0@hesh] HYDU_sock_connect (./utils/sock/sock.c:171): unable to get host address for ubuntu (1) [proxy:0:0@hesh] main (./pm/pmiserv/pmip.c:209): unable to connect to server ubuntu at port 42104 (check for firewalls!) Why I am getting there errors? Both the machines have locally generated executable hello. Do I must have same executable copied in both machines? On master node 192.168.1.4 root@ubuntu:/home# ls acme hello hosts.cfg mpi_hello.c On client 192.168.1.6 root@hesh:/home#ls hesh hello hello in each machine sis generated after compilation locally. On 22 October 2013 12:51, Jeff Hammond <jeff.science(a)gmail.com> wrote: > See http://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager > for how to launch MPI across multiple machines. You do not need to > use any special hardware adapter; any Ethernet card should provide > sufficient TCP/IP support. > > Jeff > > On Tue, Oct 22, 2013 at 9:23 AM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > Hello, > > > > You'll need to either install mpich on a shared file system (such as > NFS) or individually on each of the nodes. The README file in the mpich > tarball should have most of this information. > > > > -- Pavan > > > > On Oct 22, 2013, at 8:26 AM, heshsham basit wrote: > > > >> Hi, > >> > >> I did some experimentation with MPICH on windows, and recently came to > know that there is no support for Windows now. > >> I am now shifting to Ubuntu. I was successful in installing MPICH2 and > running my first MPI program on a single machine. > >> > >> Now I want to run my code on multiple PCs. How do I install MPI in a > LAN ? What are the additional requirement if any? Do I need any specific > hardware -network adapter My card details are as under: > >> > >> root@ubuntu:/home# lspci -nnk | grep -i net -A2 > >> 01:07.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. > RTL-8139/8139C/8139C+ [10ec:8139] (rev 10) > >> Subsystem: Hewlett-Packard Company Device [103c:301d] > >> Kernel driver in use: 8139too > >> > >> > >> Most of my machines are having this type of card and are connected in > LAN. > >> Really appreciate for any pointers to appropriate document or > suggestions. > >> > >> > >> Thanks and regards, > >> > >> > >> > >> > >> _______________________________________________ > >> discuss mailing list discuss(a)mpich.org > >> To manage subscription options or unsubscribe: > >> https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

machinefile question and select hydra as primary pm
by mark 24 Oct '13

24 Oct '13

Hello, i did a fresh install of 3.0.4. on a linux laptop and since i dont have much experience i followed the installation readme on the tar file and everything went well expect the part when i test the installation. The first test run : mpiexec -n <number> ./examples/cpi , went as planned but then the second test run on multiple nodes (mpiexec -f machinefile -n <number> ./examples/cpi) returned some errors. As the readme states there should be a machinefile. The thing is i cant find this file and tried to create it my self in the given form, just using only my laptops hostname a couple of times and placing it in the mpich installation directory ( /home/<username>/mpich-install) and that didnt fix the errors. One other thing i noticed is that on the readme states that hydra is the primary process manager, but using the mpiexec -info it clearly states that pmi ( or something similar ) is the primary pm. And for a last thing my installation was made without using any prefixes at the configuration. PS.One other thing i just noticed is that when i did the installation i could run mpich ( the test runs with the pi programm, mpiexec -info , etc) , but now after a reboot i cant run mpich on the terminal. For example *mpiexec -info* returns *_bash: mpiexec: command not found_* . Whats the deal with that? Didnt i installed mpich on my machine? Thank you for your time.

1 0

Re: [mpich-discuss] mpiexec crash
by Pavan Balaji 23 Oct '13

23 Oct '13

On Oct 23, 2013, at 5:29 PM, Cherukumilli, Vasu <Vasu_Cherukumilli(a)mentor.com> wrote: > Crash that we are seeing: > > [proxy:0:0@gretel] HYD_pmcd_pmip_control_cmd_cb (</PATH/TO>/src/pm/hydra/pm/pmiserv/pmip_cb.c:934): assert (!closed) failed > [proxy:0:0@gretel] HYDT_dmxu_poll_wait_for_event (</PATH/TO>/src/mpich2-1.5/src/pm/hydra/tools/demux/demux_poll.c:77): callback returned error status > [proxy:0:0@gretel] main (</PATH/TO>/src/mpich2-1.5/src/pm/hydra/pm/pmiserv/pmip.c:210): demux engine error waiting for event > [mpiexec@gretel] control_cb (</PATH/TO>/src/mpich2-1.5/src/pm/hydra/pm/pmiserv/pmiserv_cb.c:201): assert (!closed) failed > [mpiexec@gretel] HYDT_dmxu_poll_wait_for_event (</PATH/TO>/src/mpich2-1.5/src/pm/hydra/tools/demux/demux_poll.c:77): callback returned error status > [mpiexec@gretel] HYD_pmci_wait_for_completion (</PATH/TO>/src/mpich2-1.5/src/pm/hydra/pm/pmiserv/pmiserv_pmci.c:196): error waiting for event > [mpiexec@gretel] main (</PATH/TO>/src/mpich2-1.5/src/pm/hydra/ui/mpich/mpiexec.c:325): process manager error waiting for completion These are cleanup messages. You should have gotten an output which says so. > No such file or directory. (errno = ENOENT) This is the real error message. Did you make sure your executables are located on all the nodes in the same location? —- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0