Hi， I found that the server installed OpenMPI arms with iptables, so further communication between ib03 and ib04 is prohibited! The mpirun works fine across multi-server with hello_c
Thanks Ralph ! Thanks -Yanfei 发件人: devel [mailto:devel-boun...@open-mpi.org] 代表 Ralph Castain 发送时间: 2014年3月26日 17:45 收件人: Open MPI Developers 主题: Re: [OMPI devel] 答复: example/Hello_c.c : mpirun run failed on two physical nodes. Can you please configure OMPI with --enable-debug, and then execute mpirun -mca plm_base_verbose 10 -host ib03 hostname This will provide debug information about the problem. Thanks Ralph On Tue, Mar 25, 2014 at 9:51 PM, Wang,Yanfei(SYS) <wangyanfe...@baidu.com<mailto:wangyanfe...@baidu.com>> wrote: Hi, Thanks jeff, and I have not figured out what happened yet with this FAQ. 1. Ssh remote login OK: [root@bb-nsi-ib04 examples]# ssh ib03 hostname bb-nsi-ib03.bb01.*.com [root@bb-nsi-ib04 examples]# 2. following command return immediately without nothing returned [root@bb-nsi-ib04 examples]# mpirun -host ib03 hostname [root@bb-nsi-ib04 examples]# 3. following command excute successfully. [root@bb-nsi-ib04 examples]# ssh ib03 mpirun -------------------------------------------------------------------------- mpirun could not find anything to do. It is possible that you forgot to specify how many processes to run via the "-np" argument. -------------------------------------------------------------------------- [root@bb-nsi-ib04 examples]# So, does it seem like that the non-interactive shell profile is not correctly configured? Step 3 can execute succefully... Hope any response! BR Yanfei Wang -----邮件原件----- 发件人: devel [mailto:devel-boun...@open-mpi.org<mailto:devel-boun...@open-mpi.org>] 代表 Jeff Squyres (jsquyres) 发送时间: 2014年3月25日 22:09 收件人: Open MPI Developers 主题: Re: [OMPI devel] example/Hello_c.c : mpirun run failed on two physical nodes. Try this FAQ entry: http://www.open-mpi.org/faq/?category=running#diagnose-multi-host-problems On Mar 25, 2014, at 6:54 AM, "Wang,Yanfei(SYS)" <wangyanfe...@baidu.com<mailto:wangyanfe...@baidu.com>> wrote: > Hi, > > I am a fresh learner of OpenMPI programmer, and have some troubles on > building mpi programming, hope some helps.. > > The example/helloc_c can works successfully with 2 process on local machine, > however, do not work on two separate physical node. > > Physical two nodes: > Eg: > [root@bb-nsi-ib04 examples]# mpirun -hostfile hosts -np 2 hello_c The > command just return instantly without nothing printed. > Local machine: > [root@bb-nsi-ib04 examples]# mpirun -np 2 hello_c Hello, world, I am 0 > of 2, (Open MPI v1.7.5, package: Open MPI > root@bb-nsi-ib04.bb01.*.com<mailto:root@bb-nsi-ib04.bb01.*.com> > Distribution, ident: 1.7.5, Mar 20, 2014, 108) Hello, world, I am 1 of > 2, (Open MPI v1.7.5, package: Open MPI > root@bb-nsi-ib04.bb01.*.com<mailto:root@bb-nsi-ib04.bb01.*.com> > Distribution, ident: 1.7.5, Mar 20, 2014, 108) > [root@bb-nsi-ib04 examples]# > -----peer machine is ok-------- > [root@bb-nsi-ib03 examples]# mpirun -np 2 hello_c Hello, world, I am 0 > of 2, (Open MPI v1.7.5, package: Open MPI > root@bb-nsi-ib03.bb01.*.com<mailto:root@bb-nsi-ib03.bb01.*.com> > Distribution, ident: 1.7.5, Mar 20, 2014, 108) Hello, world, I am 1 of > 2, (Open MPI v1.7.5, package: Open MPI > root@bb-nsi-ib03.bb01.*.com<mailto:root@bb-nsi-ib03.bb01.*.com> > Distribution, ident: 1.7.5, Mar 20, 2014, 108) > [root@bb-nsi-ib03 examples]# > the command run successfully, and print two message!! > > Configuration details: > OpenMPI version: 1.7.5 > Hostfile: > [root@bb-nsi-ib04 examples]# cat hosts > ib03 slots=1 > ib04 slots=1 > [root@bb-nsi-ib04 examples]# > /etc/hosts: > [root@bb-nsi-ib04 examples]# cat /etc/hosts > 192.168.71.3 ib03 > 192.168.71.4 ib04 > SSH: > Public rsa key is redistributed two machine, ib03 and ib04, and to do ssh > login in without password is ok, I am sure. > > I am confused about this trouble, and anyone can help us? It have nothing > log and error tip, could anyone tell me how to do diagnose it. > > BR > > Yanfei Wang > > > _______________________________________________ > devel mailing list > de...@open-mpi.org<mailto:de...@open-mpi.org> > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel > Link to this post: > http://www.open-mpi.org/community/lists/devel/2014/03/14385.php -- Jeff Squyres jsquy...@cisco.com<mailto:jsquy...@cisco.com> For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ _______________________________________________ devel mailing list de...@open-mpi.org<mailto:de...@open-mpi.org> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel Link to this post: http://www.open-mpi.org/community/lists/devel/2014/03/14386.php _______________________________________________ devel mailing list de...@open-mpi.org<mailto:de...@open-mpi.org> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel Link to this post: http://www.open-mpi.org/community/lists/devel/2014/03/14396.php
