Greetings,

I have read the wiki.

My q is:  I'd like to have the root zone be on one network (inside) the
admin network tag, and
then the other physical nic is tagged with, say, prod.  And I'd like all
the vms to be there.

So, I have the following:

sandbender is the vm, and sandbooter is the GZ

root@sandbooter /usbkey]# cat /usbkey/config

coal=true
admin_nic=00:0c:29:6e:97:b1
admin_ip=198.98.0.33
admin_netmask=255.255.255.0
admin_network=
admin_gateway=198.98.0.33

prod_nic=00:0c:29:6e:97:a7
prod_ip=
prod_netmask=
prod_network=
prod_gateway=

headnode_default_gateway=198.98.0.21

dns_resolvers=8.8.8.8,8.8.4.4
dns_domain=midlant.ogt11.net

ntp_hosts=0.smartos.pool.ntp.org
compute_node_ntp_hosts=198.98.0.33

hostname=sandbooter

Here are the resultant physical nics:



And here is the json for my vm:

root@sandbooter /usbkey]# vmadm get 9f52f81b-e84d-c95a-e001-9cac52eb99a3
|json nics
[
  {
    "interface": "net0",
    "mac": "82:fb:b2:d5:ac:16",
    "nic_tag": "prod",
    "gateway": "207.188.212.1",
    "gateways": [
      "207.188.212.1"
    ],
    "netmask": "255.255.255.128",
    "ip": "207.188.212.33",
    "ips": [
      "207.188.212.33/25"
    ],
    "primary": true
  }
]

root@sandbooter /usbkey]# dladm show-phys -m
LINK         SLOT     ADDRESS            INUSE CLIENT
e1000g1      primary  0:c:29:6e:97:b1    yes  e1000g1
e1000g0      primary  0:c:29:6e:97:a7    yes  e1000g0


[root@sandbooter /usbkey]# nictagadm list
NAME           MACADDRESS         LINK           TYPE
prod           00:0c:29:6e:97:a7  e1000g0        normal
admin          00:0c:29:6e:97:b1  e1000g1        normal


snoop -d e1000g0
snoop -d e1000g1

And, finally, the traditional ifconfig shows what I want on the global zone:

[root@sandbooter /usbkey]# ifconfig -a
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232
index 1
        inet 127.0.0.1 netmask ff000000
e1000g1: flags=1001000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4,FIXEDMTU> mtu
1500 index 2
        inet 198.98.0.33 netmask ffffff00 broadcast 198.98.0.255
        ether 0:c:29:6e:97:b1
lo0: flags=2002000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv6,VIRTUAL> mtu 8252
index 1
        inet6 ::1/128



both show traffic.  Currently, both interfaces are physically connected to
the same physical switch.  Initially, sandbender was built with the nictag
"admin".  I changed that using the instructions/examples on:

https://wiki.smartos.org/display/DOC/Using+vmadm+to+manage+virtual+machines

and now it is on "prod".  Of course, it worked before the change and
doesn't now.

Again, my goal is to have the global zone on one switch and all of the VM's
public interfaces on another.
I'd of course use zlogin in an emergency.  But yes, I don't want there to
be any possibility of reaching the global zone from a vm.


Any observations?  ideas?

I'd really appreciate it.

Cheers!
-sam



-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com

Reply via email to