I got two ASG 220 units yesterday to perform some tests and decide, whether they suit our needs or not.
As part of our HA setup, we'd like to connect each ASG to one of our core switches. Using a multi-chassis environment with seperate administrative domains, we will not be able to use 802.3ad and therefore, I tried to use bonding mode 1 (active/backup).
Network design:
SWITCH 1 ==== SWITCH 2
e0/1 e0/2 e0/1 e0/2
| | | |
ASG1 ASG2 ASG1 ASG2
eth0 eth1 eth1 eth0
(pri) (bkp) (bkp) (pri)
How to change LAG mode and setting primary interface using CLI:
cc
OBJS
itfparams
link_aggregation_group
REF_LagOne[...]
mode=1
primary=REF_ItfHwDefaultInternal
w
Here is the output after having issued the changes above:
127.0.0.1 OBJS itfparams link_aggregation_group > REF_LagOne[REF_LagOne,itfparams,link_aggregation_group]
Logged into object 'REF_LagOne'. Use 'w' to write eventual changes.
{
'use_carrier' => 1,
'mode' => 1,
'status' => 1,
'primary' => 'REF_ItfHwDefaultInternal',
'updelay' => 0,
'arp_interval' => 0,
'id' => 0,
'miimon' => 0,
'enforce_mac' => 0,
'xmit_hash_policy' => 'layer2',
'name' => 'REF_LagOne',
'itfhw' => [
'REF_SwQSJNTgDn',
'REF_ItfHwDefaultInternal'
],
'comment' => '',
'ad_select' => 0,
'arp_ip_target' => '0.0.0.0',
'downdelay' => 0,
'lacp_rate' => 1
}
Long story short:
The failover simply doesn't work. Tried deactivating use_carrier and going back to MIImon (100ms) with no luck. Funny thing is, that it even ignores the primary interface (set to eth0) and instead uses eth1 as primary.
Does anyone of you guys have a running LAG in active/backup configuration?
Cheers,
gartoffel
This thread was automatically locked due to age.
