User 350 | 1/14/2015, 9:53:28 PM
I have a strange problem. We have two powerful servers: 32 cores and 256GB RAM each. When I'm running a program using: "mpiexec -n 2 -hostfile ~/machines", all the two instances are executed on one machine and I get warning: "Duplicate IP address". Once I use: "mpiexec -n 2 -pernode -hostfile ~/machines", the master node stucks on:
GRAPHLABSUBNETID/GRAPHLABSUBNETMASK environment variables not defined. Using default values Subnet ID: 0.0.0.0 Subnet Mask: 0.0.0.0 Will find first IPv4 non-loopback address matching the subnet
and the "top" on the slave node shows that the program is running 100% cpu.
The OS is "Red Hat Enterprise Linux Server release 7.0", and I tried openmpi 1.8.4 and 1.7.5.
Thanks a lot, Michael.