<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=us-ascii">
<META NAME="Generator" CONTENT="MS Exchange Server version 5.5.2654.45">
<TITLE>RE: [openib-general] OpenSM work Q1 + Apr.</TITLE>
</HEAD>
<BODY>
<P><FONT SIZE=2>> On Wed, 2005-05-04 at 13:51, Eitan Zahavi wrote:</FONT>
<BR><FONT SIZE=2>> > 1. Irresponsive port scanning during light sweep - Allow the SM</FONT>
<BR><FONT SIZE=2>> > to recognize ports that were not responding in the first sweep</FONT>
<BR><FONT SIZE=2>> > (but had a link state that is not down).</FONT>
<BR><FONT SIZE=2>> </FONT>
<BR><FONT SIZE=2>> This sounds like the bad port handling algorithm we bashed on this list.</FONT>
<BR><FONT SIZE=2>> Is it ?</FONT>
<BR><FONT SIZE=2>[EZ] It is solving exactly the opposite phenomenon:</FONT>
<BR><FONT SIZE=2>Your proposal was to solve a case (I still did not see a log showing it is really happening) where OpenSM is continuously polling a non responsive port. </FONT></P>
<P><FONT SIZE=2>While this solution is for the reverse problem where OpenSM totally ignores an Irresponsive Port after the first sweep is done - not giving it a chance to join the fabric even if it started to respond later.</FONT></P>
<BR>
<P><FONT SIZE=2>> </FONT>
<BR><FONT SIZE=2>> -- Hal</FONT>
</P>
</BODY>
</HTML>