Hi,
I have a server with two network devices: eth1 and eth2.
I have been given 4 IP addresses via a router that I don't have any control
over.
Two of them go only to eth1, and the other two only to eth2.
Each IP address has its own gateway.
The devices, ip addresses and gateways are as following:
eth1 – ip1: aaa.189.aaa.254 - GW: aaa.189.aaa.253
eth1 – ip2: bbb.143.bbb.214 - GW: bbb.143.bbb.213
eth2 – ip3: ccc.170.ccc.202 - GW: ccc.170.ccc.201
eth2 – ip4: ddd.180.ddd.101 - GW: ddd.180.ddd.100
Two or more of the IPs will be used for virtual machines.
I have set my network up in the following way:
===========================
[#] cat ifcfg-eno1
NAME=eno1
DEVICE=eno1
DEVICETYPE=TeamPort
ONBOOT=yes
TEAM_MASTER=team1
TEAM_PORT_CONFIG="{'prio': 9}"
[#] cat ifcfg-eno2
NAME=eno2
DEVICE=eno2
DEVICETYPE=TeamPort
ONBOOT=yes
TEAM_MASTER=team1
TEAM_PORT_CONFIG="{'prio': 10}"
[#] cat ifcfg-team1
BRIDGE=br0
DEVICE=team1
DEVICETYPE=Team
ONBOOT=yes
BOOTPROTO=none
TEAM_CONFIG="{'runner': {'name': 'lacp', 'active': true, 'fast_rate': true,
'tx_hash': ['eth', 'ipv4', 'ipv6']},'link_watch': {'name':
'ethtool'},"ports":{"eno1":{},"eno2":{}}}"
PROXY_METHOD=none
BROWSER_ONLY=no
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=no
NAME="Team team1"
UUID=702de3eb-2e80-897c-fd52-cd0494dd8123
[#] cat ifcfg-br0
DEVICE=br0
STP=yes
BRIDGING_OPTS=priority=32768
TYPE=Bridge
BOOTPROTO=none
#
IPADDR=aaa.189.aaa.254
PREFIX=30
GATEWAY=aaa.189.aaa.253
DNS1=aaa.bbb.ccc.aaa
DNS2=aaa.bbb.ccc.bbb
#
IPADDR1=bbb.143.bbb.214
PREFIX1=30
GATEWAY1=bbb.143.bbb.213
#
IPADDR2=ccc.170.ccc.202
PREFIX2=30
GATEWAY2=ccc.170.ccc.201
#
IPADDR3=ddd.180.ddd.101
PREFIX3=30
GATEWAY3=ddd.180.ddd.100
#
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_FAILURE_FATAL=no
NAME=br0
NM_CONTROLLED=YES
ONBOOT=yes
PROXY_METHOD=none
BROWSER_ONLY=no
UUID=d2d68553-f97e-7549-7a26-b34a26f29318"
===========================
[Question 1]
It took my some time to figure out that LACP, requires a specific setup on
the router (that I don't have control over), but the setup above was
answering for all my IP's.
I see via:
[#] teamnl team1 options
...
mode roundrobin
...
=> So is roundrobin the default mode, and takes over even when LACP is set
up in TEAM_CONFIG?
[Question 2]
=> Knowing that eth1 only answers requests to the two firsts IP addresses,
and eth2 the two last IP addresses, is mode roundrobin the only sensible
mode/runner in my case?
(Activebackup seems to only answer requests to the IPs connected to either
eth1 OR eth2, not both. Loadbalance seems have the same challenge . . . )
[Question 3]
=> If roundrobin is the most sensible mode, would it be enough to set up
'ifcfg-team1', like this?
[#] cat ifcfg-team1
...
TEAM_CONFIG='{"runner": {"name": "roundrobin"}}'
...
=> OR Would I need to set up "ports":{"eno1":{},"eno2":{}} inside the
TEAM_CONFIG?
[Question 4]
=> Does it make sense to use a team in my case, or could I just drop the
team and connected the bridge(s) to eno1 and eno2 instead?
E.g: What do I gain by setting up a team in my case?
Thanks!
Best regards
Wolf