I apparently have a sketchy internet connection: a couple times a day, my modem drops the link to my RT-AX88U for a second and picks it back up (thanks Spectrum). Any time it does this, my laptop on the wifi disconnects for 10-20 seconds.
Checking the logs, it looks like a service called wanduck restarts a whole bunch of things, and so a 1-2 second blip on the modem turns into a 5-10 second interruption. This is disruptive for streaming, zooming, etc., and so I'd like to figure out if all these service restarts are actually necessary or not.
I haven't found a good description of what wanduck does, but it seems most of the chatter on this forum is people asking to reboot everything because when their WAN goes down, they don't get DHCP or a hookup to their ISP. I'm trying to figure out the opposite -- is there a way to prevent wanduck from doing its thing, or at least not do this unless the WAN goes down for more than, say, 5 seconds? I suspect it's just a brief connection blip on the modem/ISP side and restarting everything is unnecessary.
Two typical examples from my logs:
and so on, couple times each day.
It feels like it shouldn't be necessary to reboot sambda and YazFi and all the other services every time my Spectrum cable sneezes. So anyone know if it's possible to disable this or at least put it on a longer timeout trigger? Is that even wise?
Thanks!
Checking the logs, it looks like a service called wanduck restarts a whole bunch of things, and so a 1-2 second blip on the modem turns into a 5-10 second interruption. This is disruptive for streaming, zooming, etc., and so I'd like to figure out if all these service restarts are actually necessary or not.
I haven't found a good description of what wanduck does, but it seems most of the chatter on this forum is people asking to reboot everything because when their WAN goes down, they don't get DHCP or a hookup to their ISP. I'm trying to figure out the opposite -- is there a way to prevent wanduck from doing its thing, or at least not do this unless the WAN goes down for more than, say, 5 seconds? I suspect it's just a brief connection blip on the modem/ISP side and restarting everything is unnecessary.
Two typical examples from my logs:
Code:
# (3 second outage:)
Dec 14 14:37:38 WAN(0)_Connection: WAN(0) link down.
Dec 14 14:37:38 kernel: eth0 (Int switch port: 6) (Logical Port: 6) (phyId: 13) Link Up at 2500 mbps full duplex
Dec 14 14:37:41 WAN(0)_Connection: WAN(0) link up.
# (...turns into longer outage as services cycle)
Dec 14 14:37:41 rc_service: wanduck 1488:notify_rc restart_wan_if 0
Dec 14 14:37:41 custom_script: Running /jffs/scripts/service-event (args: restart wan_if)
Dec 14 14:37:41 lldpd[2380]: removal request for address of XXX.XXX.XXX.XXX, but no knowledge of it
Dec 14 14:37:41 dnsmasq[6350]: read /etc/hosts - 22 names
Dec 14 14:37:41 dnsmasq[6350]: using nameserver XXX.XXX.XXX.XXX
...
Dec 14 14:37:43 wan: finish adding multi routes
...
Dec 14 14:37:48 rc_service: udhcpc_wan 26954:notify_rc stop_samba
Dec 14 14:37:48 rc_service: udhcpc_wan 26954:notify_rc start_samba
....
Dec 14 14:37:53 YazFi: YazFi v4.4.4 starting up
Dec 14 14:38:03 YazFi: Forcing YazFi Guest WiFi clients to reauthenticate
...
Code:
Dec 14 13:12:14 kernel: eth0 (Int switch port: 6) (Logical Port: 6) (phyId: 13) Link DOWN.
Dec 14 13:12:19 kernel: eth0 (Int switch port: 6) (Logical Port: 6) (phyId: 13) Link Up at 2500 mbps full duplex
Dec 14 13:12:19 WAN(0)_Connection: WAN(0) link up.
Dec 14 13:12:19 rc_service: wanduck 1488:notify_rc restart_wan_if 0
Dec 14 13:12:19 custom_script: Running /jffs/scripts/service-event (args: restart wan_if)
... (20 seconds later...)
Dec 14 13:12:41 YazFi: Forcing YazFi Guest WiFi clients to reauthenticateand so on, couple times each day.
It feels like it shouldn't be necessary to reboot sambda and YazFi and all the other services every time my Spectrum cable sneezes. So anyone know if it's possible to disable this or at least put it on a longer timeout trigger? Is that even wise?
Thanks!
