I had the pi-hole setup successfully for awhile, using all the standard install implementations, plus configuring for DNS over HTTPS. I went to check the admin page for the first time in weeks and couldnt pull it up.. Same with SSH to the IPv4 address.
Expected Behaviour:
Be able to access the admin page from 192.168.1.158/admin and SSH in from the same IP on port 22.
Actual Behaviour:
Admin page inaccessible. Same with SSH. IP is static'd in router and ping, but connections via putty are refused. I'm able to SSH in using raspberrypi.local and running ifconfig shows no IPv4 address is assigned to eth0, just an iIPv6 which appears to be a link-local address - inet6 fe80::f1a0:dad5:44d1:2837
pihole -d indicated no IPv4 assigned and that it doesn't match what's found in /etc/pihole/setupVars.conf
/etc/dhcpcd.conf is showing the correct values in it as well. I'm not entirely sure why it decided to start picking up IPv6 instead of v4 but I'm at a loss.
I'd attach a token, but the debug log won't upload. If needed I can copy/paste the log into here.
This is a network configuration issue, not directly related to Pi-hole.
It could be that your IPv4 address lease has expired and your Pi-hole machine failed to renew it, but then again that should not happen with static IP addresses if configured correctly on-device.
Another cause might be that your network changed, e.g. from 192.168.1.0/24 to 192.168.42.0/24, which could also lead to your Pi-hole being inaccessible under its previous static address.
However, that wouldn't agree with your observation of your Pi-hole machine having no IPv4 address assigned at all.
Let's see if we can spot some obvious errors by taking a quick look at the output of the following commands (run on your Pi-hole machine):
That's what I was kind of figuring - that the issue is residing on the network config somewhere and not the pi-hole given it's nature.. I can confirm the subnet for the home network hasn't changed. as for the leases, the pi-hole was setup with a statically reserved IP both in the pihole setup page and on the router's reservation table, neither of which have been messed with since initial setup.
ip -4 address show results:
Blockquote
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
I missed your answer - sorry for getting back late.
Your dhcpcd.conf looks perfectly normal - there is nothing in it that would explain why your device fails to register IPv4 addresses for your eth0 network interface as shown by ip -4 address.
If it wouldn't be for your IPv6 link-local address working on that interface, I would suspect some kind of error with your eth0 hardware.
Another guess: If you would have upgraded your system's OS from a very old Raspbian (Jessie and before), there may be remnants of old network configurations interfering with dhcpcd. You would find those in /etc/network/interfaces (so I consider this highly unlikely, as you stated it had worked for some time before).
As said before: This is a network configuration issue.
There are no hints at Pi-hole being involved in this.
I’m running this off of a raspberry pi 4 I bought back in April on the most recent version of raspbian stretch. Right now it’s hooked into an asus ac56u Router running the latest version of dd-wrt.
At this point I’m more curious why it did this than anything else. I’ve got an r410 and an 520 server I just got set up as virtual hosts that I was thinking of running pi hole off of as a virtual machine.
If you are still lacking IPv4 addresses: RPi power issues are infamous to cause the most bizarre error situations.
So as a precaution, you could check for signs of recent under voltage issues by running
grep -i "voltage" /var/log/syslog
Other than that, I am out of ideas for what could cause an IPv4 address failing to register on a network interface while an IPv6 link-local succeeds.
As this isn't Pi-hole related, you may want to consider other sources of support as well (if you haven't already done so).
