Solved upgrading blade enclosure firmware to version 1.20!<br><br>Thanks.<br><br><div><span class="gmail_quote">On 18 Jan 2007 09:01:45 -0500, <b class="gmail_sendername">Hal Rosenstock</b> <<a href="mailto:halr@voltaire.com">
halr@voltaire.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">On Thu, 2007-01-18 at 08:52, Davide Pasquale wrote:<br>
> On 18 Jan 2007 08:19:34 -0500, Hal Rosenstock <<a href="mailto:halr@voltaire.com">halr@voltaire.com</a>><br>> wrote:<br>> On Thu, 2007-01-18 at 08:02, Davide Pasquale wrote:<br>> ><br>
> > On 18 Jan 2007 07:34:43 -0500, Hal Rosenstock<br>> <<a href="mailto:halr@voltaire.com">halr@voltaire.com</a>><br>> > wrote:<br>> > On Thu, 2007-01-18 at 06:19, Davide Pasquale wrote:
<br>> > > Starting opensm I see this error in<br>> /var/log/osm.log:<br>> > ><br>> > > OpenSM Rev:openib-2.0.5 OpenIB svn Exported<br>> revision
<br>> > > Jan 18 12:11:39 628147 [95AA8160] -><br>> osm_vendor_bind:<br>> > Binding to port<br>> > > 0x18feffff8c7a8d<br>> > > Jan 18 12:11:39 629557 [95AA8160] ->
<br>> osm_vendor_bind:<br>> > Binding to port<br>> > > 0x18feffff8c7a8d<br>> > > Jan 18 12:11:39 630605 [41401960] -> SM port is<br>> down
<br>> > > Jan 18 12:11:39 630693 [41401960] -><br>> > __osm_sm_state_mgr_signal_error:<br>> > > ERR 3207: Invalid signal OSM_SM_SIGNAL_DISCOVER in
<br>> state<br>> > > IB_SMINFO_STATE_DISCOVERING<br>> > > Jan 18 12:11:49 631170 [41E02960] -> SM port is<br>> down<br>> > > Jan 18 12:11:49 631238 [41E02960] ->
<br>> > __osm_sm_state_mgr_signal_error:<br>> > > ERR 3207: Invalid signal OSM_SM_SIGNAL_DISCOVER in<br>> state<br>> > > IB_SMINFO_STATE_DISCOVERING
<br>> > ><br>> > > and the SM port is always down.<br>> ><br>> > The error message is benign.<br>> ><br>> > Is the SM port plugged into any other IB device ?
<br>> ><br>> > -- Hal<br>> ><br>> > Hi Hal,<br>> ><br>> > we are using HP Blade System and each blade has an<br>> infiniband card
<br>> > onboard.<br>> > The SM port is plugged in the Infiniband switch internal to<br>> the blade<br>> > enclosure.<br>> > Is this information helpful for you ?
<br>><br>> The port being down has nothing to do with SM operation. For<br>> some<br>> reason, there is no connectivity or negotiation between the<br>> blades and<br>> the switch.
<br>><br>> -- Hal<br>><br>> ><br>> ><br>> ><br>> ><br>> ><br>><br>><br>> Thanks!<br>> What can I look to in order to solve this problem ?
<br><br>I don't know the HP blade system so the only thing I can say to try is<br>to unseat and reseat all the blades (HCAs and switch(es)) to see if this<br>resolves the problem. If it doesn't, I have no clue.<br>
<br>-- Hal<br><br>><br>> Regards,<br>> Davide.<br>><br>><br><br></blockquote></div><br>