With v2 SI , the VMI got updated so that the VM gets the SI virtual IP itself during DHCP
Affects | Status | Importance | Assigned to | Milestone | ||
---|---|---|---|---|---|---|
Juniper Openstack | Status tracked in Trunk | |||||
R3.0 |
Fix Committed
|
High
|
Hari Prasad Killi | |||
Trunk |
Fix Committed
|
High
|
Hari Prasad Killi |
Bug Description
R3.0 2715 Ubuntu 14.04 Kilo multi-node
vn1 : 10.1.1.0/24
vn2 : 20.1.1.0/24
VMs : 10.1.1.3 and 20.1.1.3 were created as left and right endpoint vms
Created a port 10.1.1.100 , 20.1.1.100 in vn1, vn2 and attached it to a v2 in-network nat SI
A VM was booted with these two ports as left and right interfaces.
On the left intf (eth0), VM got 10.1.1.100 rightly through DHCP
But on the right intf (eth1), VM got 20.1.1.4 (which is the SI VIP on the right VN)
Repeated the same with new set of ports (10.1.1.101, 20.1.1.101) , (10.1.1.102, 20.1.1.102 ) etc.. Here, the vms consistently got the expected IPs (i.e. they did not get VIP)
Looking at the port-list, the port 89e3551f-
root@anushka-
+------
| id | name | mac_address | fixed_ips |
+------
| bc9f7df7-
| | | | {"subnet_id": "9af28776-
| 75c5efe2-
| c7fb412c-
| 285a3ad7-
| | | | {"subnet_id": "45744f89-
| 419c056f-
| | | | {"subnet_id": "45744f89-
| e1baa1ae-
| | | | {"subnet_id": "45744f89-
| 89e3551f-
| | | | {"subnet_id": "9af28776-
| afbd34ae-
| | | | {"subnet_id": "9af28776-
+------
root@anushka-
root@anushka-
% Total % Received % Xferd Average Speed Time Time Time Current
100 4524 100 4524 0 0 520k 0 --:--:-- --:--:-- --:--:-- 552k
{
"virtual-
"fq_name": [
],
"href": "http://
"id_perms": {
},
"uuid": {
}
},
{
],
},
{
],
}
],
"name": "89e3551f-
"perms2": {
},
{
],
}
],
{
},
],
}
],
{
],
}
],
"uuid": "89e3551f-
{
},
{
},
{
}
]
},
]
},
},
{
],
}
],
{
],
}
],
{
},
{
},
{
}
]
}
}
}
Changed in juniperopenstack: | |
milestone: | r3.0-fcs → future |
tags: | added: blocker |
agent team needs to fix this. qualify to make sure that service chain ip flag is not set in instance ip when responding to dhcp