Todo #2397
closed
Gateway Groups
100%
Description
Currently when defining a new Gateway group, the default trigger level is set to 'Member Down'.
It ideally should be set to 'Packet Loss or High Latency' as 'Member Down' will only cause a failover when a member interface has lost its link. Which might not happen in most WAN environments.
Updated by Seth Mos about 13 years ago
Make sure to touch gwlb.inc return_gateways_array(); it defaults to memberdown for any dynamic gateway. That needs to be changed there.
Updated by Warren Baker about 13 years ago
- Status changed from New to Feedback
- % Done changed from 0 to 100
Applied in changeset 1cc71979e44d7955084a0cdb50d7698239fac770.
Updated by Chris Buechler about 13 years ago
- Category set to Gateways
"member down" doesn't mean link down, or it never has historically up to and including recent 2.1, it means it's completely down (100% loss) in apinger. Packet loss or high latency is more suitable in some instances, but I'm not sure that should be the default. That will lead to flapping for some users, especially where under load they normally have high latency and some packet loss (which comprises probably a slight majority of all users). I think that's probably too touchy of a default.
Updated by Warren Baker about 13 years ago
Hrmm. I think http://doc.pfsense.org/index.php/Multi-WAN_2.0#Trigger_Level needs to be changed then to indicate this as it specifically mentions that a member needs to lose its link before been marked as down.
Updated by Warren Baker about 13 years ago
Applied in changeset 8de4a8bc4d52755dce1fbf2fe80d45687397a429.
Updated by Warren Baker about 13 years ago
- Status changed from Feedback to Rejected
Currently when defining a new Gateway group, the default trigger level is set to 'Member Down'.
It ideally should be set to 'Packet Loss or High Latency' as 'Member Down' will only cause a failover when a member interface has lost its link. Which might not happen in most WAN environments.