<meta http-equiv="Content-Type" content="text/html; charset=utf-8"><div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Feb 7, 2017 at 2:08 PM, Dmitriy Lyubimov <span dir="ltr"><<a href="mailto:dlieu.7@gmail.com" target="_blank">dlieu.7@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote"><span class="gmail-">On Tue, Feb 7, 2017 at 11:15 AM, Balaji, Pavan <span dir="ltr"><<a href="mailto:balaji@anl.gov" target="_blank">balaji@anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div>
<div class="gmail-m_-189295826531197626m_-6411825982758373274BodyFragment"><font size="2"><span style="font-size:10pt">
<div class="gmail-m_-189295826531197626m_-6411825982758373274PlainText">Hi Dmitriy,<br>
<br>
You should use publish/lookup name for this, instead of relying on manual printing. I've attached updated server and client codes that do that. Please see attached.<br>
<br>
You'll need to start the Hydra nameserver for this using:<br>
<br>
% hydra_nameserver<br>
<br>
Once the nameserver has started, you can connect all mpiexecs to it using something like:<br>
<br>
% mpiexec -nameserver localhost ./server 2<br></div></span></font></div></div></blockquote><div><br></div></span><div>Thanks but we have our own name resolution architecture (which is not manual printing/pasting). This is simple example that's been asked for to isolated the problem and provided, not the actual code. </div><div><br></div><div>The name is guaranteed to be delivered to client verbatim down to the bit. The use is fully consistent with the MPI 3 spec. </div></div></div></div></blockquote><div><br></div><div>Sorry for reiterating on this point. Just checked again, here's the quote from 3.1 MPI standard, MPI_COMM_Connect():</div><div><br></div><div>"port_name is the address of the server. It must be the same as the name returned
by MPI_OPEN_PORT on the server."</div><div><br></div><div>So, the standard is very clear about it. </div><div><br></div><div>For as long as the port name is faithfully communicated from process that opened it to every other process, it does not have to be obtained via any strictly prescribed means of redistribution.</div><div><br></div><div>And any error in the port name should just raise MPI_ERR_PORT. Which is not what we observe, so I am afraid using or not using MPI_Lookup_name is not the problem / a solution of this issue.</div><div><br></div><div> </div></div></div></div>