I'm running the pi-hole docker on my QNAP NAS. It has a managing app for docker images, which uses some complex virtual switch setup.
I'm not a docker expert.
I can access the internet, websites, etc.
Clients are OSX, iPhone, iPads
Some websites load extremely slow, and I expect this to be because some timeouts have to happen before things move on, but I don't have an idea what to do about it.
pi-hole doesn't use IP6 name-servers (it's disabled, so I can't enable these, and I don't know why). My internal network is using IP4, but there might be some IP6 stuff going on, I'm not aware of.
Does anyone have an idea what to check, look at, try out to get rid of these delays?
| => nmap --reason pihole -p443 -Pn
Starting Nmap 7.70 ( https://nmap.org ) at 2019-07-25 16:51 CEST
Nmap scan report for pihole (192.168.1.100)
Host is up, received user-set (0.0068s latency).
rDNS record for 192.168.1.100: pihole.home
PORT STATE SERVICE REASON
443/tcp closed https conn-refused
Nmap done: 1 IP address (1 host up) scanned in 29.06 seconds
________________________________________________________________________________
| ~/Desktop @ imac-pro-559 (robby)
| => nmap --reason pihole -p80 -Pn
Starting Nmap 7.70 ( https://nmap.org ) at 2019-07-25 16:52 CEST
Nmap scan report for pihole (192.168.1.100)
Host is up, received user-set (0.0058s latency).
rDNS record for 192.168.1.100: pihole.home
PORT STATE SERVICE REASON
80/tcp open http syn-ack
Nmap done: 1 IP address (1 host up) scanned in 21.04 seconds
________________________________________________________________________________
*** [ DIAGNOSING ]: Networking
[✓] IPv4 address(es) bound to the eth0 interface:
192.168.1.100/24 does not match the IP found in /etc/pihole/setupVars.conf (https://discourse.pi-hole.net/t/use-ipv6-ula-addresses-for-pi-hole/2127)
[✗] No IPv6 address(es) found on the eth0 interface.
Hmm... seems like things start to become slow again... I see a very constant level of queries, even at night, where I would expect much less queries. And I don't understand the clients level... I hope I'm not running 10.000 clients somewhere I'm not aware of.
All the requests are coming from my router, which uses pi-hole as its DNS server. And the pi-hole IP is not accessible from the outside. So, I don't think it's an open relay. Is there any way to test this?
Since around 10:00 today the queries dropped down from about 12.000 in an hour to 200. I think I restarted the container... hell I'm getting old, I can't remember
The top domains are DNS Discovery Service requests. With this number of requests, you may have a loop between the router and Pi-Hole. This can be caused by conditional forwarding.
@amilroy I set the external DNS of my router manually to the pi-hole.
@jfb I enabled spanning-tree protocol on my router but seems it didn't detect it. And yes, I had "conditional forwarding" enabled and disabled it, which might have triggered the restart and the drop of the number of queries.
Response time is much better now. There are only occasionally delays and some apps on my iPhone or iPad show dome delays but looks good so far.