<br><br>
<div><span class="gmail_quote">On 7/26/07, <b class="gmail_sendername">Eitan Zahavi</b> <<a href="mailto:eitan@mellanox.co.il">eitan@mellanox.co.il</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div>
<div><span><font face="Palatino Linotype" color="#0000ff"><strong>I propose that when there is no MTU in the partition policy file OpenSM use a </strong></font></span></div>
<div><span><font face="Palatino Linotype" color="#0000ff"><strong>configurable default from: </strong></font></span><span><strong><font face="Palatino Linotype" color="#0000ff">/etc/cache/opensm/opensm.opt.</font></strong>
</span></div></div></blockquote>
<div> </div>
<div>That would make this the default rather than 2K. IMO it should be when some "special" unused mtu is set in the partition config.</div>
<div> </div>
<div>-- Hal</div><br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div>
<div><span><strong><font face="Palatino Linotype" color="#0000ff">Something like:</font></strong></span></div>
<div><span><strong><font face="Palatino Linotype" color="#0000ff"># The default MTU to be used for IPoIB and other MCGs when the partition-policy </font></strong></span></div>
<div><span><strong><font face="Palatino Linotype" color="#0000ff"># does not provide exact value. The default is the lowest possible MTU</font></strong></span></div>
<div><span><strong><font face="Palatino Linotype" color="#0000ff">mcg_default_mtu 1</font></strong></span></div><span class="q">
<div><span><strong><font face="Palatino Linotype" color="#0000ff"></font></strong></span> </div>
<div><span lang="en-gb"><b><i><font face="Monotype Corsiva" color="#0000ff" size="6">Eitan Zahavi</font></i></b><i></i></span> <br><span lang="en-gb"><font face="Tahoma" size="2">Senior Engineering Director, Software Architect
</font></span> <br><span lang="en-gb"><font face="Tahoma" size="2">Mellanox Technologies LTD</font></span> <br><span lang="en-gb"><font face="Tahoma" size="2">Tel:+972-4-9097208<br>Fax:+972-4-9593245</font></span> <br><span lang="en-gb">
<font face="Tahoma" size="2">P.O. Box 586 Yokneam 20692 ISRAEL</font></span> </div>
<div> </div><br></span>
<blockquote style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px solid; MARGIN-RIGHT: 0px">
<div lang="en-us" dir="ltr" align="left">
<hr>
<font face="Tahoma" size="2"><b>From:</b> Shirley Ma [mailto:<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:xma@us.ibm.com" target="_blank">xma@us.ibm.com</a>] <br><b>Sent:</b> Wednesday, July 25, 2007 10:45 PM
<br><b>To:</b> Eitan Zahavi<br><b>Cc:</b> <a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:general@lists.openfabrics.org" target="_blank">general@lists.openfabrics.org</a>; Hal Rosenstock
<div><span class="e" id="q_114011bdbae4e42f_3"><br><b>Subject:</b> RE: [ofa-general] Re: openSM: Different IB MTUs<br></span></div></font><br> </div>
<div><span class="e" id="q_114011bdbae4e42f_5">
<div></div>
<p>Hello Eitan, Hal,<br><br>Thanks. It's good openSM has the configuration option to set up these attributes in MC. Is this a good idea to add below to openSM: When there is no MTU defined in the configuration file, SM can pick up the smallest link MTU in the fabrics by default? MTU is unlikely rate, slower rate might indicate the cablling problem. So using the smallest link MTU in the fabrics might not be a bad choice for MC by default. The reason I request here is to create IP multicast group, MTU is not an attribute of the group. When mapping IP multicast to IB multicast, IB muliticast might fail because of different IB link MTU size in the group, but IP multicast group will be successful without knowing the failure. If admin sets MTU in configuration file, admin would know this failure. Otherwise, admin/users could spend too much time on debugging their broken multicasting applications.
<br><br>Thanks<br>Shirley Ma<br><br><img height="16" alt="Inactive hide details for "Eitan Zahavi" <eitan@mellanox.co.il>" src="cid:896424205@26072007-1411" width="16" border="0">"Eitan Zahavi" <
<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:eitan@mellanox.co.il" target="_blank">eitan@mellanox.co.il</a>><br><br><br>
<table cellspacing="0" cellpadding="0" width="100%" border="0">
<tbody>
<tr valign="top">
<td width="40%">
<ul>
<ul>
<ul>
<ul><b><font size="2">"Eitan Zahavi" <<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:eitan@mellanox.co.il" target="_blank">eitan@mellanox.co.il</a>></font></b><font size="2"> </font>
<p><font size="2">07/25/07 12:25 PM</font></p></ul></ul></ul></ul></td>
<td width="60%">
<table cellspacing="0" cellpadding="0" width="100%" border="0">
<tbody>
<tr valign="top">
<td width="1%"><img height="1" alt="" src="cid:896424205@26072007-1418" width="58" border="0"><br>
<div align="right"><font size="2">To</font></div></td>
<td width="100%"><img height="1" alt="" src="cid:896424205@26072007-1418" width="1" border="0"><br><font size="2">"Hal Rosenstock" <<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:hal.rosenstock@gmail.com" target="_blank">
hal.rosenstock@gmail.com</a>>, Shirley Ma/Beaverton/IBM@IBMUS</font></td></tr>
<tr valign="top">
<td width="1%"><img height="1" alt="" src="cid:896424205@26072007-1418" width="58" border="0"><br>
<div align="right"><font size="2">cc</font></div></td>
<td width="100%"><img height="1" alt="" src="cid:896424205@26072007-1418" width="1" border="0"><br><font size="2"><<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:general@lists.openfabrics.org" target="_blank">
general@lists.openfabrics.org</a>></font></td></tr>
<tr valign="top">
<td width="1%"><img height="1" alt="" src="cid:896424205@26072007-1418" width="58" border="0"><br>
<div align="right"><font size="2">Subject</font></div></td>
<td width="100%"><img height="1" alt="" src="cid:896424205@26072007-1418" width="1" border="0"><br><font size="2">RE: [ofa-general] Re: openSM: Different IB MTUs</font></td></tr></tbody></table>
<table cellspacing="0" cellpadding="0" border="0">
<tbody>
<tr valign="top">
<td width="58"><img height="1" alt="" src="cid:896424205@26072007-1418" width="1" border="0"></td>
<td width="336"><img height="1" alt="" src="cid:896424205@26072007-1418" width="1" border="0"></td></tr></tbody></table></td></tr></tbody></table><br><b><font color="#0000ff" size="4">Hi Shirley,</font></b><br><font size="4">
</font><br><b><font color="#0000ff" size="4">I think I understand where your question comes from...</font></b><br><b><font color="#0000ff" size="4">Many have issue with heterogonous fabrics where not all nodes have same MTU or Speed.
</font></b><br><b><font color="#0000ff" size="4">Especially when IPoIB relies on all nodes joining the broadcast group.</font></b><br><font size="4"></font><br><b><font color="#0000ff" size="4">The term "join" for multicast groups is a little overloaded.
</font></b><br><b><font color="#0000ff" size="4">If a node joins an existing MC group it has to have a rate (speed * width) > MCG.rate and support MTU > MCG.MTU otherwise it is denied.</font></b><br><b><font color="#0000ff" size="4">
If the join is actually a "create" the node has to provide the rate and MTU which define the MCG values.</font></b><br><font size="4"></font><br><b><font color="#0000ff" size="4">To allow for administrator to control the IPoIB MCGs MTU and rate OpenSM provides the means to control these
</font></b><br><b><font color="#0000ff" size="4">values per partition. See the doc/partition-config.doc</font></b><font size="4"> </font><br><b><font color="#0000ff" size="4">Still the administrator should know what would be the lowest MTU and rate the nodes expected to join the IPoIB subnet have.
</font></b><br><b><font color="#0000ff" size="4">The tradeoff is in the hands of the administrator who can set a value that will prevent slow nodes from joining the group, </font></b><br><b><font color="#0000ff" size="4">
or assign a low value that will fit all nodes but slow down communication ...</font></b><br><font size="4"></font><br><b><font color="#0000ff" size="4">EZ</font></b>
<p><b><i><font color="#0000ff" size="7">Eitan Zahavi</font></i></b><font size="4"> </font><br>Senior Engineering Director, Software Architect<font size="4"> </font><br>Mellanox Technologies LTD<font size="4"> </font><br>
Tel:+972-4-9097208<br>Fax:+972-4-9593245<font size="4"> </font><br>P.O. Box 586 Yokneam 20692 ISRAEL<font size="4"> </font>
<p><font size="4"></font><br><br>
<hr align="left" width="100%" size="2">
<b>From:</b> <a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:general-bounces@lists.openfabrics.org" target="_blank">general-bounces@lists.openfabrics.org</a> [<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:general-bounces@lists.openfabrics.org" target="_blank">
mailto:general-bounces@lists.openfabrics.org</a>] <b>On Behalf Of </b>Hal Rosenstock<b><br>Sent:</b> Wednesday, July 25, 2007 10:01 PM<b><br>To:</b> Shirley Ma<b><br>Cc:</b> <a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:general@lists.openfabrics.org" target="_blank">
general@lists.openfabrics.org</a><b><br>Subject:</b> [ofa-general] Re: openSM: Different IB MTUs<font size="4"><br></font><br><font size="4">Shirley,<br></font><br><font size="4">On 7/25/07, </font><b><font size="4">Shirley Ma
</font></b><font size="4"> <</font><a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:xma@us.ibm.com" target="_blank"><u><font color="#0000ff" size="4">xma@us.ibm.com</font></u></a><font size="4">> wrote:
</font>
<ul><font size="4">Hal,<br><br>Thanks for your prompt reply. I am asking for how openSM handle different link MTUs in SA MCMemberRecord MTU. For example, if we have some links MTU as 2K, some links MTU as 1K. Then when enabling IPoIB, how does SM decide IPoIB broadcast group MCMemberRecord MTU size? When creating an IB multicast group from a 2K MTU node first, which PMTU value is attaching to this IB multicast group MCMemberRecord MTU?
</font></ul><br><font size="4"></font><br><font size="4">MCMemberRecord MTU gets the group MTU (when created). This is either this first joiner with sufficient components or preconfigured (and MTU can be set in the config). If a joiner has insufficient MTU for the group, it is denied.
</font><br><font size="4"></font><br><font size="4">-- Hal</font><br><font size="4"></font><br>
<ul><font size="4">Thanks<br>Shirley Ma<br><br></font><img height="16" alt="Inactive hide details for "Hal Rosenstock" <hal.rosenstock@gmail.com>" src="cid:896424205@26072007-141F" width="16"><font size="4">
"Hal Rosenstock" < </font><a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:hal.rosenstock@gmail.com" target="_blank"><u><font color="#0000ff" size="4">hal.rosenstock@gmail.com</font></u>
</a><font size="4">><br><br></font>
<table cellspacing="0" cellpadding="0" width="100%" border="0">
<tbody>
<tr valign="top">
<td width="54%">
<ul>
<ul>
<ul>
<ul>
<ul>
<ul>
<ul>
<ul><b>"Hal Rosenstock" <</b><a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:hal.rosenstock@gmail.com" target="_blank"><b><u><font color="#0000ff">hal.rosenstock@gmail.com</font></u></b>
</a><b>></b>
<p>07/25/07 10:57 AM</p></ul></ul></ul></ul></ul></ul></ul></ul></td>
<td width="46%">
<table cellspacing="0" cellpadding="0" width="100%" border="0">
<tbody>
<tr valign="top">
<td width="22%"><img height="1" src="cid:896424205@26072007-1426" width="58">
<div align="right">To</div></td>
<td width="78%"><img height="1" src="cid:896424205@26072007-1426" width="1"><br>Shirley Ma/Beaverton/IBM@IBMUS</td></tr>
<tr valign="top">
<td width="22%"><img height="1" src="cid:896424205@26072007-1426" width="58">
<div align="right">cc</div></td>
<td width="78%"><img height="1" src="cid:896424205@26072007-1426" width="1"><u><font color="#0000ff"><br></font></u><a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:general@lists.openfabrics.org" target="_blank">
<u><font color="#0000ff">general@lists.openfabrics.org</font></u></a></td></tr>
<tr valign="top">
<td width="22%"><img height="1" src="cid:896424205@26072007-1426" width="58">
<div align="right">Subject</div></td>
<td width="78%"><img height="1" src="cid:896424205@26072007-1426" width="1"><br>Re: openSM: Different IB MTUs</td></tr></tbody></table>
<table cellspacing="0" cellpadding="0" width="100%" border="0">
<tbody>
<tr valign="top">
<td width="15%"><img height="1" src="cid:896424205@26072007-1426" width="1"></td>
<td width="85%"><img height="1" src="cid:896424205@26072007-1426" width="1"></td></tr></tbody></table></td></tr></tbody></table><font size="5"><br>Shirley,</font><font size="4"><br></font><font size="5"><br>On 7/25/07, </font>
<b><font size="5">Shirley Ma</font></b><font size="5"> <</font><a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:xma@us.ibm.com" target="_blank"><u><font color="#0000ff" size="4"> </font></u><u><font color="#0000ff" size="5">
xma@us.ibm.com</font></u></a><font size="5">> wrote: </font>
<ul>
<ul><font size="5">Hello Hal,<br><br>How does openSM handle CAs with different MTUs in the same subnet? For example, IPoIB broadcast group MTU, IB multicast group PMTU? Does openSM pick up the smallest MTU in the subnet?
</font></ul></ul><font size="4"><br></font><font size="5"><br>Are you asking about link MTU, SA PathRecord/MultiPathRecord MTU, SA MCMemberRecord MTU, or all of these ?</font><font size="4"><br></font><font size="5"><br>-- Hal
</font>
<ul>
<ul><font size="5">Thanks<br>Shirley Ma</font></ul></ul><br></ul><br>
<p></p>
<p></p>
<p></p></p></p></p></span></div></blockquote></div></blockquote></div><br>