SC 3.2 - scinstall fails cluster transport discovery

Question

Hello,
I'm trying to install a cluster on 2 v240 servers running Solaris 10 U4 (08/07) built from a custom Jumpstart profile. Both systems have 4 network connections (2 public, 2 private), spread across a total of 3 VLANs (one public, two private). The cluster software installs just fine, but when it comes time to run 'scinstall' to build the cluster, I get the following error:
  Cluster Creation
    Log file - /var/cluster/logs/install/scinstall.log.4152
    Checking installation status ... done
    The Sun Cluster software is installed on "fusion03".
    The Sun Cluster software is installed on "fusion04".
    Starting discovery of the cluster transport configuration.
    Probes were sent out from all transport adapters configured for this
    node ("fusion03"). But, they were not seen by any of the other nodes.
    This may be due to any number of reasons, including improper cabling
    or a switch which was confused by the probes.
    You can either attempt to correct the problem and try the probes again
    or manually configure the transport. To correct the problem might
    involve re-cabling, changing the configuration, or fixing hardware.
    You must configure the transport manually to configure tagged VLAN
    adapters and non tagged VLAN adapters on the same private interconnect
    VLAN.
    Do you want to try again (yes/no) [yes]?I've been poking at this all day, so any pointers would be greatly appreciated.
fpsm

Answer 1

I can plumb and configure an IP on all of interfaces in both machines and they can see each other (ping and arp both work fine).
I did find one thing while looking through the log though. I am configuring these through SSH, and it appears that the SSH banner was causing a bit of a problem. Notice in the output below that the expected probe information is the words from the legal disclaimer in the SSH banner.
===========================
fusion03
===========================
scrconf -n cmd=discover_send,adapters=bge2:bge3,vlans=0:0,token=suncluster_fusion-dmz,sendcount=30
===========================
fusion04
===========================
ssh root@fusion04 -o "BatchMode yes" -n "/bin/sh -c '/usr/cluster/lib/scadmin/lib/cmd_autodiscovery 0:0 suncluster_fusion-dmz 2 30; /bin/echo SC_COMMAND_STATUS=\$?'"
quit
SC_COMMAND_STATUS=0
All users of this system have consented to, and are subject to, the
provisions of Corporate Policy Regarding Electronic Communications.
This system is for the use of authorized users only.  Individuals
using this computer system without authority, or in excess of their
authority, are subject to having all of their activities on this
system monitored and recorded by systems personnel.  In the course
of monitoring individuals improperly using this system, or in the
course of system maintenance, the activities of authorized users may
also be monitored.  Anyone using this system expressly consents to
such monitoring and is advised that if such monitoring reveals possible
criminal activity, system personnel may provide the evidence of such
monitoring to law enforcement officials.
===========================
"fusion04" found an expected probe from "of".
"fusion04" found an expected probe from "Corporate".
"fusion04" found an expected probe from "is".
"fusion04" found an expected probe from "computer".
"fusion04" found an expected probe from "subject".
"fusion04" found an expected probe from "and".
"fusion04" found an expected probe from "individuals".
"fusion04" found an expected probe from "system".
"fusion04" found an expected probe from "monitored.".
"fusion04" found an expected probe from "and".
"fusion04" found an expected probe from "system".
"fusion04" found an expected probe from "law".
    Probes were sent out from all transport adapters configured for this
    node ("fusion03"). But, they were not seen by any of the other nodes.
    This may be due to any number of reasons, including improper cabling
    or a switch which was confused by the probes.
    You can either attempt to correct the problem and try the probes again
    or manually configure the transport. To correct the problem might
    involve re-cabling, changing the configuration, or fixing hardware.
    You must configure the transport manually to configure tagged VLAN
    adapters and non tagged VLAN adapters on the same private interconnect
    VLAN.I removed the banner and those messages stopped appearing in the log, but it didn't resolve the problem - I can't get the interconnect to work.
fpsm