Multiwan GW check too fast after interface has obtained an ip?
-
When I refresh the IP on one of the WAN interfaces, I notice this in the logs. Basically apinger is brought up and within a second, the gateway status is checked and of course, the gateway status could not be determined. Does there need to be some delay for apinger to gather enough pings to figure out the health of the link?
Feb 4 18:37:20 php: : Gateways status could not be determined, considering all as up/active.
Feb 4 18:37:20 php: : Gateways status could not be determined, considering all as up/active.
Feb 4 18:37:20 php: : Gateways status could not be determined, considering all as up/active.
Feb 4 18:37:19 php: : phpDynDNS: No change in my IP address and/or 25 days has not passed. Not updating dynamic DNS entry.
Feb 4 18:37:19 php: : DynDns: Cached IP: x
Feb 4 18:37:19 php: : DynDns: Current WAN IP: x
Feb 4 18:37:19 php: : DynDns debug information: x extracted from local system.
Feb 4 18:37:19 php: : DynDns: _checkIP() starting.
Feb 4 18:37:19 php: : DynDns: _detectChange() starting.
Feb 4 18:37:19 php: : DynDns: updatedns() starting
Feb 4 18:37:19 php: : DynDns: Running updatedns()
Feb 4 18:37:19 apinger: Starting Alarm Pinger, apinger(33576) -
Once apinger gets status, it'll kick off a filter reload if any of them are down. If they're all up, it already has them all up. It works as it should, though in the case of an IP change if you do have a down gateway it'll consider it up for 1-2 seconds. That's less than ideal but happens under very rare circumstances and changing that is one of those things that's likely to cause a lot of unintended consequences.