[
https://issues.apache.org/jira/browse/CLOUDSTACK-9717?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15876415#comment-15876415
]
ASF GitHub Bot commented on CLOUDSTACK-9717:
--------------------------------------------
Github user remibergsma commented on the issue:
https://github.com/apache/cloudstack/pull/1878
@sureshanaparti why do the mac addresses need to be the same on both
routers? We're also executing arpings to update our neighbours. Networking wise
there is no need for them to be the same. I've seen it on other parts of the
code as well and I really wonder why we do this.
> [VMware] RVRs have mismatching MAC addresses for extra public NICs
> ------------------------------------------------------------------
>
> Key: CLOUDSTACK-9717
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-9717
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Components: Network Controller, VMware
> Reporter: Suresh Kumar Anaparti
> Assignee: Suresh Kumar Anaparti
> Fix For: 4.10.0.0
>
>
> [CLOUDSTACK-985|https://issues.apache.org/jira/browse/CLOUDSTACK-985] doesn't
> seem to be completely fixed.
> ISSUE
> ==================
> If there are two public networks on two VLANs, and a pair redundant VRs
> acquire IPs from both, the associated NICs on the redundant VRs will have
> mismatching MAC addresses.
> The example below shows the eth2 NICs for the first public network
> (210.140.168.0/21) have matching MAC addresses (06:c4:b6:00:03:df) as
> expected, but the eth3 NICs for the second one (210.140.160.0/21) have
> mismatching MACs (02:00:50:e1:6c:cd versus 02:00:5a:e6:6c:d5).
> *r-43584-VM (Master)*
> 6: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UNKNOWN
> qlen 1000
> link/ether 06:c4:b6:00:03:df brd ff:ff:ff:ff:ff:ff
> inet 210.140.168.42/21 brd 210.140.175.255 scope global eth2
> inet 210.140.168.20/21 brd 210.140.175.255 scope global secondary eth2
> 8: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UNKNOWN
> qlen 1000
> link/ether 02:00:50:e1:6c:cd brd ff:ff:ff:ff:ff:ff
> inet 210.140.162.124/21 brd 210.140.167.255 scope global eth3
> inet 210.140.163.36/21 brd 210.140.167.255 scope global secondary eth3
> *r-43585-VM (Backup)*
> 6: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
> link/ether 06:c4:b6:00:03:df brd ff:ff:ff:ff:ff:ff
> inet 210.140.168.42/21 brd 210.140.175.255 scope global eth2
> inet 210.140.168.20/21 brd 210.140.175.255 scope global secondary eth2
> 8: eth3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
> link/ether 02:00:5a:e6:6c:d5 brd ff:ff:ff:ff:ff:ff
> inet 210.140.162.124/21 brd 210.140.167.255 scope global eth3
> inet 210.140.163.36/21 brd 210.140.167.255 scope global secondary eth3
> CloudStack should ensure that the NICs for all public networks have matching
> MACs.
> REPRO STEPS
> ==================
> 1) Set up redundant VR.
> 2) Set up multiple public networks on different VLANs.
> 3) Acquire IPs in the RVR network until the VRs get IPs in the different
> public networks.
> 4) Confirm the mismatching MAC addresses.
> EXPECTED BEHAVIOR
> ==================
> Redundant VRs have matching MACs for all public networks.
> ACTUAL BEHAVIOR
> ==================
> Redundant VRs have matching MACs only for the first public network.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
