<br><font size=2 face="sans-serif">Hi </font>
<br><font size=2 face="sans-serif"> Lei</font>
<br><font size=2 face="sans-serif">This is working .</font>
<br><font size=2 face="sans-serif"> MVAPICH2 is supposed to
detect the active port automatically , why is it not working ...??</font>
<br>
<br>
<br>
<br>
<br>
<table width=100%>
<tr valign=top>
<td width=40%><font size=1 face="sans-serif"><b>LEI CHAI <chai.15@osu.edu></b>
</font>
<p><font size=1 face="sans-serif">06/18/2008 03:03 AM</font>
<td width=59%>
<table width=100%>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">To</font></div>
<td><font size=1 face="sans-serif">biswajit@crlindia.com</font>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">cc</font></div>
<td><font size=1 face="sans-serif">mvapich-discuss@cse.ohio-state.edu</font>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">Subject</font></div>
<td><font size=1 face="sans-serif">Re: [mvapich-discuss] problem with running
mvapich</font></table>
<br>
<table>
<tr valign=top>
<td>
<td></table>
<br></table>
<br>
<br>
<br><font size=3>Hi,<br>
<br>
MVAPICH2 is supposed to detect the active port automatically for you. Could
you try the following options:<br>
<br>
$ mpiexec -n 2 -env MV2_IBA_HCA mthca1 -env MV2_DEFAULT_PORT 1 ./a.out<br>
<br>
and see if it works for you?<br>
<br>
Lei<br>
<br>
<br>
----- Original Message -----<br>
From: biswajit@crlindia.com<br>
Date: Tuesday, June 17, 2008 6:55 am<br>
Subject: [mvapich-discuss] problem with running mvapich<br>
To: mvapich-discuss@cse.ohio-state.edu<br>
<br>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif">When I ran a simple MPI application
with mvapich2-1.0.2, I got the following error messages:</font><font size=3>
<br>
</font><font size=2 face="sans-serif"><i><br>
</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>Unknown Mellanox PCI-Express
HCA best guess as Mellanox PCI-Express SDR</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>[3] Abort: Not enough ports
are in active stateneeded active ports 1</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i> at line 424 in file
rdma_iba_priv.c</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>rank 3 in job 1 n23_32790
caused collective abort of all ranks</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i> exit status of rank
3: return code 252</i></font><font size=3> <br>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif">But there is a active port in
each node. See the below <i>'ibstat' </i>output.</font><font size=3> <br>
<br>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>CA 'mthca0'</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
CA type: MT25204</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Number of ports: 1</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Firmware version: 1.1.0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Hardware version: a0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Node GUID: 0x0019bbfffff70cb8</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
System image GUID: 0x0019bbfffff70cbb</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Port 1:</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
State: Down</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Physical state: Polling</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Rate: 10</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Base lid: 0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
LMC: 0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
SM lid: 0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Capability mask: 0x02510a68</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Port GUID: 0x0019bbfffff70cb9</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>CA 'mthca1'</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
CA type: MT25204</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Number of ports: 1</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Firmware version: 1.1.0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Hardware version: a0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Node GUID: 0x0019bbfffff7fbe8</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
System image GUID: 0x0019bbfffff7fbeb</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Port 1:</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
State: Active</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Physical state: LinkUp</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Rate: 20</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Base lid: 226</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
LMC: 0</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
SM lid: 117</i></font><font size=3> </font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Capability mask: 0x02510a68</i></font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"><i>
Port GUID: 0x0019bbfffff7fbe9</i></font><font size=3>
<br>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif"> And, whenever I run same
job in nodes with IB port 1 active, it works properly.</font><font size=3>
</font><font size=2 face="sans-serif"><br>
> </font><font size=2 face="sans-serif">Is there any option in MVAPICH
to select the IB port which should be used ?</font><font size=3> <br>
<br>
> _______________________________________________<br>
> mvapich-discuss mailing list<br>
> mvapich-discuss@cse.ohio-state.edu<br>
> </font><a href="http://mail.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss"><font size=3>http://mail.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss</font></a><font size=3>
</font>
<br>