Hi All, in the document http://www.linux-ha.org/IpFailoverChannelBonding a possible concern for HA is indicated as a disagreement with "standby" mode which I can't totally see.
There is mentioned that the active-backup mode defines the transmit policy only whereas receiving is always in round robin mode. This is not described in the Linux Ethernet Bonding Driver HOWTO at all, that's true, but I can't even get how this statement could be observed. According to the Bonding HOWTO, in active-backup mode only one slave in the bond is active and a different slave becomes active if, and only if, the active slave fails. Actually, as far as I experienced this is eventually true. The active-backup mode enables one enslaved interface only and "disables" all the others. And this should be true for receiving policy too, also because the NOARP option is given for each enslaved interfaces which would exclude these to be reached from incoming packets (maybe this is the same in round robin mode, never tried it). Anyway, the active-backup mode, as the name suggests, sounds to be a bit more logical with this kind of configuration rather than a configuration where basically fault tolerance is provided for outgoing packets and load balancing for incoming ones. Jan _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
