Use this procedure to check multicast groups for proper
membership.
To check multicast groups for proper membership, perform
the following procedure:
- If you are running a host-based subnet manager, to check
multicast group creation, on the fabric management server run a query
for the specific subnet. Remember that you must provide
the HCA and port through which the subnet manager connects to the subnet.
/sbin/saquery –o mcmember –h [HCA] –p [HCA port]
Each
interface produces an entry like the following example. Note the 4 KB
maximum transmission unit (MTU) and 20 g rate.
GID: 0xff12601bffff0000:0x0000000000000016
PortGid: 0xfe80000000000000:0002550070010f00
MLID: 0xc004 PKey: 0xffff Mtu: 4096 Rate: 20g PktLifeTime: 2147 ms
QKey: 0x00000000 SL: 0 FlowLabel: 0x00000 HopLimit: 0xff TClass: 0x00
Note: You
can check for misconfigured interfaces by using something like the following
example, which looks for any MTU that is not 4096 or rate is 10 g:
/sbin/saquery –o mcmember –h [HCA] –p [port] | egrep –B 3 –A 1 ‘Mtu: [0-
3]|Rate: 10g'
- If you are running an embedded subnet manager, to check
multicast group creation, run the following command on each switch
with a master subnet manager. If you have set it up, you
might use the dsh command from the Cluster Systems
Management/Management Server (CSM/MS) to the switches. For details,
see Setting up remote command processing. Remember to
use --devicetype IBSwitch::Qlogic when pointing
to the switches.
smShowGroups
There should
be just one group with all the HCA devices on the subnet being part
of the group.
Note: MTU=5 indicates 4 KB. MTU=4 indicates 2 KB.
The
following example shows 4 KB MTU.
0xff12401bffff0000:00000000ffffffff (c000)
qKey = 0x00000000 pKey = 0xFFFF mtu = 5 rate = 3 life = 19 sl = 0
0x00025500101a3300 F 0x00025500101a3100 F 0x00025500101a8300 F
0x00025500101a8100 F 0x00025500101a6300 F 0x00025500101a6100 F
0x0002550010194000 F 0x0002550010193e00 F 0x00066a00facade01 F