Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hello, I using a bonded interface for drbd and corosync. Each interface is connected to a different network switch, therefore I'll use mode 1 (active-backup). Sometimes a switch is blocking traffic but is still connected on the link layer. When using miimon to check for active connection a blocked switch is not detected. Therefore I'm using ARP monitoring and I few IP targets beyond the directly connected switches. My question is: What is a good value for arp_interval? I'm using 500ms. Are there some best practice values? This is a snipped of my config .... # For the bonding failover, arp monitoring is used. # Do not use the directly connected switch IP addresses # to the network interface but devices beyond the # directly connected switches. # use IP address on the same subnet as the bonding interface! # ON MER1 USE : vmhost1, san1a, san2a, san3a #BONDING_OPTS="mode=1 arp_interval=500 arp_validate=all arp_ip_target=172.16.111.10,172.16.111.20,172.16.111.23,172.16.111.29" # ON MER2 USE: vmhost1, san1b, san2b, san3b BONDING_OPTS="mode=1 arp_interval=500 arp_validate=all arp_ip_target=172.16.111.10,172.16.111.21,172.16.111.24,172.16.111.30" TYPE=Bonding ... ---------------------------------------------------------------------------------------------------------- Best Regards, Maurits van de Lande Van de Lande BV Raamsdonksveer The Netherlands -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20130320/d75a7478/attachment.htm>