<html><head><meta http-equiv="Content-Type" content="text/html; charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><br class=""><div><br class=""><blockquote type="cite" class=""><div class="">On May 9, 2020, at 2:24 AM, Martin Ivanov <<a href="mailto:marto1980@gmail.com" class="">marto1980@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class="">Hello Tony,<div class="">Thank you very much for your reply. I followed your advice and gave the hostname 'dragonfly' to my machine. Then, in /etc/hosts I provided gave the alias 'dragonfly' to localhost as you suggested:</div><div class=""><br class=""></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">"</span></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">marto@dragonfly% cat /etc/hosts</span><br class=""></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">::1                     localhost dragonfly
</span><br class="">127.0.0.1               localhost dragonfly<br class=""></span></div><div class=""><span style="font-family:monospace" class="">"</span></div><div class=""><br class=""></div><div class="">I compiled both the icpi and hellow examples. Now mpiexec seems to work, although not as reliably as I might wish. Running any of the examples with one core like this:<br class=""></div><div class=""><br class=""></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">mpiexec -n 1 mpich-3.0.4/examples/hellow</span><br class=""></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class=""><br class=""></span></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">is always successful. I could never get the command to finish with 2, 3, or 4 cores: it simply hangs. E.g. with 2 cores after running the above command I get:</span></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class=""><br class=""></span></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">"</span></span></div><div class=""><span style="font-family:monospace" class=""><span style="" class="">marto@dragonfly% ps x | grep hellow
</span><br class="">166493 ??  I6s      0:00.00 mpich-3.0.4/examples/hellow
<br class="">166494 ??  I6s      0:00.00 mpich-3.0.4/examples/hellow<br class=""></span></div><div class=""><span style="font-family:monospace" class="">"</span></div><div class=""><span style="font-family:monospace" class=""><span style="" class=""><br class=""></span></span></div></div></div></blockquote><div><br class=""></div><div>Ok, good, that seemed like it was the obvious problem.  Can you show all the processes involved in the launch, not just the application (i.e. also the mpirun and anything that is spawned)?  This might provide a further clue about what is happening underneath.</div><br class=""></div><div>Tony</div><div><br class=""></div></body></html>