I am running some tests on a HA setup where a device A can only reach a device B through one of the switches. In reality there will be a link for device b to both switches, but I am just testing the scenario where device b is only reachable through one of the switches.

device A ---- switch1 --- device b
       |         |
       |         |
       |----- switch2

Device A is running debian and its using the bonding driver in active-backup mode for the 2 eth ports. The link is monitored by pointing the arp_ip_target to device B.

Disconnecting any of the cables works as expected, but in the case of disconnecting the cable between the switches it recurringly logs this in syslog every 5 seconds (instead of only stating once that the link is down):

[ 5638.977739] bond0: (slave eth1): link status definitely up
[ 5639.307739] bond0: (slave eth1): link status definitely down, disabling slave

Functionally it works as expected (switches to use eth0), but would be good to know why it thinks the link is temporarily up + if there is a way to avoid it (as it spams the logs like that).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.