<br><br><div class="gmail_quote">On Fri, Jun 7, 2013 at 6:52 PM, Orion Poplawski <span dir="ltr"><<a href="mailto:orion@cora.nwra.com" target="_blank">orion@cora.nwra.com</a>></span> wrote:<br><blockquote style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid" class="gmail_quote">
<div class="im">On 06/07/2013 02:23 PM, Hal Rosenstock wrote:<br>
<br>
</div><div class="im"><blockquote style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid" class="gmail_quote">
Also, if you turn on log verbosity on OpenSM temporarily and send me the log<br>
for that run, I could see what is going on with in terms of trying to set the<br>
non default subnet prefix with the Windows node. Given the log you sent, I can<br>
only imagine that the SMA on the Windows node is ack'ing the PortInfo set<br>
which sets the subnet prefix but not really acting on it properly.<br>
-- Hal<br>
</blockquote>
<br></div>
Full log is at <a href="http://sw.cora.nwra.com/test/opensm.debug.log.gz" target="_blank">http://sw.cora.nwra.com/test/<u></u>opensm.debug.log.gz</a><br>
<br></blockquote><div> </div><div>Looking at that log, I didn't see _any_ MC joins from that port (GUID 0x5ad00000c5ced) so this is a different scenario than before :-(</div><div> </div><div>Also, the previous confusion with:</div>
<div> </div><div># saquery -m 0xc000<br> PortGid.................fe80::<u></u>1:5:ad00:c:5c3d (Topspin DDR-HCAe LX x8)<div class="im"><br> PortGid.................fe80::<u></u>1:19:bbff:ff00:5851 (saga mthca0)<br>
</div> PortGid.................fe80::<u></u>1:19:bbff:ff00:3899 (sfcomp1 mthca0)<div class="im"><br> PortGid.................fe80::<u></u>1:1a:4bff:ff0c:20c9 (HP Lion Cub 128MB)<br> <font color="#ff0000">PortGid.................fe80::<u></u>5:ad00:c:5ced (MT25204 InfiniHostLx Mellanox Technologies)</font><br>
PortGid.................fe80::<u></u>1:17:8ff:ffd0:9df9 (alexandria2 HCA-1)<br></div><div class="im">GUID is <font color="#ff0000">5:ad00:c:5ced</font> and prefix is <font color="#ff0000">fe80::<u></u></font> so it's either missing a digit like 1 (fe80::1 like the others) or if it's a 0 it would have a 3rd colon (fe80:::). So I'm not sure what's going on there either.</div>
<div class="im"> </div></div><div> </div><div> </div><blockquote style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid" class="gmail_quote">
I had fontdb shutdown when I started opensm - then booted it up.<br> </blockquote><blockquote style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid" class="gmail_quote">
<br>
This seems to be when it first comes up (lid 0, prefix 0xfe80::0)<br>
<br>
Jun 07 14:56:58 088453 [193D0700] 0x10 -> osm_pi_rcv_process: [<br>
Jun 07 14:56:58 088465 [193D0700] 0x08 -> PortInfo dump:<br>
port number..............1<br>
node_guid................<u></u>0x0005ad00000c5cec<br>
port_guid................<u></u>0x0005ad00000c5ced<br>
m_key....................<u></u>0x0000000000000000<br>
subnet_prefix............<u></u>0xfe80000000000000<br>
base_lid.................0<br>
master_sm_base_lid.......0<br>
capability_mask..........<u></u>0x2500A68<br>
diag_code................0x0<br>
m_key_lease_period.......0x0<br>
local_port_num...........1<br>
link_width_enabled.......0x3<br>
link_width_supported.....0x3<br>
link_width_active........0x2<br>
link_speed_supported.....0x3<br>
port_state...............INIT<br>
state_info2..............0x52<br>
m_key_protect_bits.......0x0<br>
lmc......................0x0<br>
link_speed...............0x13<br>
mtu_smsl.................0x20<br>
vl_cap_init_type.........0x30<br>
vl_high_limit............0x0<br>
vl_arb_high_cap..........0x8<br>
vl_arb_low_cap...........0x8<br>
init_rep_mtu_cap.........0x4<br>
vl_stall_life............0xFF<br>
vl_enforce...............0x30<br>
m_key_violations.........0x0<br>
p_key_violations.........0x0<br>
q_key_violations.........0x0<br>
guid_cap.................0x20<br>
client_reregister........0x0<br>
mcast_pkey_trap_suppr....0x0<br>
subnet_timeout...........0x0<br>
resp_time_value..........0x10<br>
error_threshold..........0xF0<br>
max_credit_hint..........0x0<br>
link_round_trip_latency..0x0<br>
capability_mask2.........0x0<br>
link_speed_ext_active....0x0<br>
link_speed_ext_supported.0x0<br>
link_speed_ext_enabled...0x0<br>
Jun 07 14:56:58 088495 [193D0700] 0x08 -> Capability Mask:<br>
IB_PORT_CAP_HAS_TRAP<br>
IB_PORT_CAP_HAS_AUTO_MIG<br>
IB_PORT_CAP_HAS_SL_MAP<br>
IB_PORT_CAP_HAS_LED_INFO<br>
IB_PORT_CAP_HAS_SYS_IMG_GUID<br>
IB_PORT_CAP_HAS_VEND_CLS<br>
IB_PORT_CAP_HAS_CAP_NTC<br>
IB_PORT_CAP_HAS_CLIENT_REREG<br>
Jun 07 14:56:58 088499 [193D0700] 0x04 -> osm_pi_rcv_process: Discovered port num 1 with GUID 0x5ad00000c5ced for parent node GUID 0x5ad00000c5cec, TID 0x130e<br>
<br>
<br>
Then later, sm seems to have assigned a lid.<br>
<br>
Jun 07 14:56:58 090679 [161CB700] 0x08 -> PortInfo dump:<br>
port number..............1<br>
node_guid................<u></u>0x0005ad00000c5cec<br>
port_guid................<u></u>0x0005ad00000c5ced<br>
m_key....................<u></u>0x0000000000000000<br>
subnet_prefix............<u></u>0xfe80000000000001<br>
base_lid.................16<br>
master_sm_base_lid.......1<br>
capability_mask..........<u></u>0x2500A68<br>
diag_code................0x0<br>
m_key_lease_period.......0x0<br>
local_port_num...........1<br>
link_width_enabled.......0x3<br>
link_width_supported.....0x3<br>
link_width_active........0x2<br>
link_speed_supported.....0x3<br>
port_state...............INIT<br>
state_info2..............0x52<br>
m_key_protect_bits.......0x0<br>
lmc......................0x0<br>
link_speed...............0x13<br>
mtu_smsl.................0x40<br>
vl_cap_init_type.........0x30<br>
vl_high_limit............0x0<br>
vl_arb_high_cap..........0x8<br>
vl_arb_low_cap...........0x8<br>
init_rep_mtu_cap.........0x4<br>
vl_stall_life............0xFF<br>
vl_enforce...............0x30<br>
m_key_violations.........0x0<br>
p_key_violations.........0x0<br>
q_key_violations.........0x0<br>
guid_cap.................0x20<br>
client_reregister........0x1<br>
mcast_pkey_trap_suppr....0x0<br>
subnet_timeout...........0x12<br>
resp_time_value..........0x10<br>
error_threshold..........0x88<br>
max_credit_hint..........0x0<br>
link_round_trip_latency..0x0<br>
capability_mask2.........0x0<br>
link_speed_ext_active....0x0<br>
link_speed_ext_supported.0x0<br>
link_speed_ext_enabled...0x0<br>
Jun 07 14:56:58 090709 [161CB700] 0x08 -> Capability Mask:<br>
IB_PORT_CAP_HAS_TRAP<br>
IB_PORT_CAP_HAS_AUTO_MIG<br>
IB_PORT_CAP_HAS_SL_MAP<br>
IB_PORT_CAP_HAS_LED_INFO<br>
IB_PORT_CAP_HAS_SYS_IMG_GUID<br>
IB_PORT_CAP_HAS_VEND_CLS<br>
IB_PORT_CAP_HAS_CAP_NTC<br>
IB_PORT_CAP_HAS_CLIENT_REREG<br>
Jun 07 14:56:58 090713 [161CB700] 0x08 -> osm_pi_rcv_process: Client reregister received on response<br>
Jun 07 14:56:58 091294 [12FC6700] 0x10 -> osm_db_store: ]<br>
Jun 07 14:56:58 091301 [12FC6700] 0x10 -> osm_lid_mgr_process_subnet: ]<br>
Jun 07 14:56:58 091308 [161CB700] 0x10 -> pi_rcv_process_set: [<br>
Jun 07 14:56:58 091313 [161CB700] 0x08 -> pi_rcv_process_set: Received logical SetResp() for GUID 0x5ad00000c5ced, port num 1<br>
for parent node GUID 0x5ad00000c5cec TID 0x1311<br>
Jun 07 14:56:58 091320 [161CB700] 0x08 -> osm_db_update: Key:0x0005ad00000c5ced previously exists in:/var/cache/opensm/guid2mkey with value:0x0000000000000000<br>
Jun 07 14:56:58 091324 [161CB700] 0x10 -> pi_rcv_process_set: ]<br>
Jun 07 14:56:58 091327 [161CB700] 0x10 -> osm_pi_rcv_process: ]<br>
<br>
But I'm not really sure what I'm looking for.<div class="HOEnZb"><div class="h5"><br>
<br>
-- <br>
Orion Poplawski<br>
Technical Manager <a href="tel:303-415-9701%20x222" target="_blank" value="+13034159701">303-415-9701 x222</a><br>
NWRA, Boulder/CoRA Office FAX: <a href="tel:303-415-9702" target="_blank" value="+13034159702">303-415-9702</a><br>
3380 Mitchell Lane <a href="mailto:orion@nwra.com" target="_blank">orion@nwra.com</a><br>
Boulder, CO 80301 <a href="http://www.nwra.com" target="_blank">http://www.nwra.com</a><br>
</div></div></blockquote></div><br>