Possible bug in v4.15.2-1

adam:ONE®
1

Since late Friday UTC i’ve been seeing a DNS issue which caused failure which stopped a slave device monitoring the UPS connected to my pfSense gateway.

To cut a long story short on all my servers on my LAN I am seeing:

root@hostd:~# nslookup -q=A gateway.howitts.co.uk
Server:         127.0.0.53
Address:        127.0.0.53#53

Non-authoritative answer:
Name:   gateway.howitts.co.uk
Address: 216.106.178.113
Name:   gateway.howitts.co.uk
Address: 172.17.2.254
Name:   gateway.howitts.co.uk
Address: 172.17.2.253

root@hostd:~# nslookup -q=A pfsense
Server:         127.0.0.53
Address:        127.0.0.53#53

Non-authoritative answer:
Name:   pfsense.howitts.co.uk
Address: 216.106.178.113
Name:   pfsense.howitts.co.uk
Address: 172.17.2.254

pfsense is not in the hosts file but is forced to resolve to my LAN IP by pfSense. gateway.howitts.co.uk is in the hosts file and similarly resolves to my LAN IP, 172.17.2.254.

I have absolutely no idea where the results 216.106.178.113 and 172.17.2.253 are coming from. The 216.106.178.113 broke the UPS monitor, apcupsd, running on one of my servers as it kept spamming me with messages that it had lost contact with the router. I have changed its config to use the IP address for the moment and it is working fine.

I can see in the dashboard that adamOne is using the hosts file. I just don’t understand the results:

image
image1299×147 3.06 KB

and:

image
It is only doing this for the pfSense gateway. All other readings of the hosts file seem to be correct.

2

My memory has pointed me to how the .253 address came into being. See How can I get mytools.management to work with HAproxy . In the end, for that problem I used a workaround of not using a split DNS for the webserver, but in getting there I initially went down the alias route, hence 172.17.2.253. So adamOne has a long memory? How can I make it forget? It is not in /usr/local/etc/adamone/anmuscle.conf or anywhere else in pfSense.

Also it does not explain the 261 address at all.

3

Thanks for this report @NickH – I have just created a support ticket to gather log level 6 diagnostics from you so we can investigate.