2

I have a Mac OS Machine that I've set up a LACP connection with a cisco SG300. 

bond0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
    options=23<RXCSUM,TXCSUM,TSO4>
    ether 00:0a:cd:26:c3:f6 
    inet6 fe80::20a:cdff:fe26:c3f6%bond0 prefixlen 64 scopeid 0x9 
    inet 192.168.16.10 netmask 0xffffff00 broadcast 192.168.16.255
    nd6 options=1<PERFORMNUD>
    media: autoselect (1000baseT <full-duplex,flow-control,energy-efficient-ethernet>)
    status: active
    bond interfaces: en8 en6

en6: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500
    options=23<RXCSUM,TXCSUM,TSO4>
    ether 00:0a:cd:26:c3:e3 
    media: autoselect (1000baseT <full-duplex,flow-control,energy-efficient-ethernet>)
    status: active
en8: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500
    options=23<RXCSUM,TXCSUM,TSO4>
    ether 00:0a:cd:26:c3:f6 
    media: autoselect (1000baseT <full-duplex,flow-control,energy-efficient-ethernet>)
    status: active

When using the lag connection I can only ping certain IP addresses for some strange reason.

When I disable the lag connection I can ping all addresses fine. I can also ping remote IP addresses fine when connected. 

8.8.8.8       yes - google
192.168.16.2  no - vsphere vm (dns)
192.168.16.28 yes - vsphere vm
192.168.16.3  no - physical device (backup dns)

I see no correlation between why some addresses respond and others do not. I've tried disabling all the firewalls. Of course all these IP's respond when using a different connection.

I've tried arp -a. it shows these ips on the network.

? (192.168.16.1) at 0:50:56:b3:7d:54 on bond0 ifscope
? (192.168.16.2) at 0:50:56:b3:e:37 on bond0 ifscope
? (192.168.16.3) at b8:27:eb:7a:b8:c1 on bond0 ifscope
? (192.168.16.28) at 0:50:56:b3:52:b0 on bond0 ifscope
? (192.168.16.50) at 0:50:56:b3:e:8a on bond0 ifscope

One other thing I noticed is that the connection can ping all addresses if I remove one of the connections.

tried clearing arp cache. tried restarting mac and cisco.

Improve this question

1 Answer 1

Reset to default
0

If only some remote hosts are working and not others, it's likely there's something wrong with the LACP either on Linux or on the switch.

Only the traffic which hashes to one slave is working, which is why some hosts work and some don't.

On Linux, check /proc/net/bonding/bondX and make sure you have the same Aggregator ID on both ports.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .