(Multiwan) Short hickup when Tier2 goes down

[shadesh]

Howdy Folks!

I have a good running MultiWANSetup (Tier1 VDSL and Tier2 LTE) running at my homeoffice but i noticed some strange behavior when the Tier2 WAN members goes down (LTE in germany has some strange forced separation after 24h). My internet connectivity goes down for just a few seconds.

system.log:

Code: [Select]

Apr  7 12:55:00 gw opnsense: /usr/local/etc/rc.filter_configure: Ignore down inet gateways : LTE_PPP
Apr  7 12:55:00 gw opnsense: /usr/local/etc/rc.filter_configure: ROUTING: keeping current default gateway '62.155.*.*'
Apr  7 12:55:00 gw opnsense: /usr/local/etc/rc.filter_configure: Ignore down inet6 gateways : LTE_PPP
Apr  7 12:55:00 gw opnsense: /usr/local/etc/rc.filter_configure: ROUTING: keeping current default gateway 'fe80::86b5:*:*:*%pppoe0'
Apr  7 12:55:03 gw kernel: pflog0: promiscuous mode disabled
Apr  7 12:55:03 gw kernel: pflog0: promiscuous mode enabled
Apr  7 12:55:56 gw opnsense: /usr/local/etc/rc.filter_configure: ROUTING: keeping current default gateway '62.155.*.*'
Apr  7 12:55:56 gw opnsense: /usr/local/etc/rc.filter_configure: ROUTING: keeping current default gateway 'fe80::86b5:*:*:*%pppoe0'
gateway.log:

Code: [Select]

Apr  7 12:54:59 gw dpinger: LTE_PPP 8.8.4.4: Alarm latency 702994us stddev 1282031us loss 0%
Apr  7 12:54:59 gw dpinger: GATEWAY ALARM: LTE_PPP (Addr: 8.8.4.4 Alarm: 1 RTT: 702994ms RTTd: 1282031ms Loss: 0%)
Apr  7 12:55:55 gw dpinger: LTE_PPP 8.8.4.4: Clear latency 391019us stddev 790004us loss 1%
Apr  7 12:55:55 gw dpinger: GATEWAY ALARM: LTE_PPP (Addr: 8.8.4.4 Alarm: 0 RTT: 391019ms RTTd: 790004ms Loss: 1%)

For me it looks like that OPNSense is flushing the state table and recreating it multiple times, one time after Tier2 goes down and one time when Tier2 is coming back online. Is there a way to ignore the state change on the non active Tier2 gateway?

Thanks in advance!

[shadesh]

Nobody with a problem like that?