I have an ASUS ZenWiFi BQ16 Pro (x2) with firmware updated to 3.0.0.6.102_36998 on both the router and node.
I've noticed a very strange, consistent behavior that started about two weeks ago, which I'm trying to debug. What happens is that, once a day, I see the following message in my system log, after which the Internet becomes completely inaccessible:
The reason I suspect this line is that I'm using Smokeping to monitor ICMP latencies, and this event always coincides with the five-minute block when Internet connectivity is lost. It's always present in the log during the five-minute window when the outage occurs, and it only appears during a window when an outage happens. There seems to be a strong correlation: this message appears right as the Internet cuts out, and the Internet only cuts out when this message is logged. I've even had it happen mid-Zoom call, so I was able to narrow it down to the minute.
When the Internet is out, I can ping the router and the IP addresses of other machines on my local network, but I cannot ping 209.51.188.17 (mail.gnu.org). However, if I ssh into the router, I can ping 209.51.188.17, but I cannot resolve mail.gnu.org.
I've discovered two things happening on the router:
1. The
2. There is no
Has anyone seen this before? Are there any settings you know of that might cause both my NAT and DNS configurations to get wiped after a DHCP renewal with my ISP? This is a very odd yet very consistent behavior that I've only ever seen before with this BQ16 Pro router, and only recently.
Currently, I'm disabling services one by one, trying to simplify my environment as much as possible. The only recent change I've made to my local network—around the time this started happening—was setting up a Technitium DNS Server on one of my machines and using it as the DNS resolver for all local equipment. However, thinking that the ISP might not like me no longer using their DNS servers, I changed it so that two noisy machines still use the ISP's DNS. This did not resolve the issue.
Thank you,
John
¯
I've noticed a very strange, consistent behavior that started about two weeks ago, which I'm trying to debug. What happens is that, once a day, I see the following message in my system log, after which the Internet becomes completely inaccessible:
Sep 6 15:18:33 dhcp client: renew <IP>/255.255.255.0 via <IP> for 300 seconds.The reason I suspect this line is that I'm using Smokeping to monitor ICMP latencies, and this event always coincides with the five-minute block when Internet connectivity is lost. It's always present in the log during the five-minute window when the outage occurs, and it only appears during a window when an outage happens. There seems to be a strong correlation: this message appears right as the Internet cuts out, and the Internet only cuts out when this message is logged. I've even had it happen mid-Zoom call, so I was able to narrow it down to the minute.
When the Internet is out, I can ping the router and the IP addresses of other machines on my local network, but I cannot ping 209.51.188.17 (mail.gnu.org). However, if I ssh into the router, I can ping 209.51.188.17, but I cannot resolve mail.gnu.org.
I've discovered two things happening on the router:
1. The
/etc/resolv.conf file is always empty after the DHCP client renewal. If I manually add "nameserver 9.9.9.9" to this file, the router can resolve DNS names again.2. There is no
MASQUERADE rule in the iptables -t nat table. If I add one manually using iptables -t nat -A POSTROUTING -s 192.168.1.0/24 -o eth1 -j MASQUERADE, then all the machines on my local network can access the Internet again, as if nothing had happened.Has anyone seen this before? Are there any settings you know of that might cause both my NAT and DNS configurations to get wiped after a DHCP renewal with my ISP? This is a very odd yet very consistent behavior that I've only ever seen before with this BQ16 Pro router, and only recently.
Currently, I'm disabling services one by one, trying to simplify my environment as much as possible. The only recent change I've made to my local network—around the time this started happening—was setting up a Technitium DNS Server on one of my machines and using it as the DNS resolver for all local equipment. However, thinking that the ISP might not like me no longer using their DNS servers, I changed it so that two noisy machines still use the ISP's DNS. This did not resolve the issue.
Thank you,
John
¯
