I played a detective last night.
What preceded it:
- A few hours before that, I ordered an upgrade of my primary FFTx over the Internet
- 300Mbps upload at the cost of the original 150Mbps
- Two hours later, there was a short WAN outage because the profile was modified at the ISP. I expected it, and I also received an SMS. Provisioning was done.
- after about one minute, everything was running.
What I discovered a few hours later:
- I couldn't access my services over the WAN
- the same services were available via LAN
- step by step Network controller - OK
- Nginx proxy manager, Pi-hole, internal DNS, ... - OK
- primary NAS, where all services are available from LAN - OK
I started to get nervous. And hungry. Two things you shouldn't meet in a man.
Bingo! The last of the tests where I used my public fixed IP address and port 80 found the problem. Surprisingly, I looked at the landing page of the ISP's router.
Reason:
When changing a subscriber's profile, it is crucial to consider the possible settings during the provisioning process of the customer's order (ISP side). ONT BRIDGE mode was to be one of the preserved parameters (when it was used before). Unfortunately, I probably discovered a hole in the ISP's provisioning system. So, yes, the bridge mode hasn't been preserved.
Logically, my own router was in DHCP WAN mode, so everything ran from my LAN in terms of services without the need for NAT (normal browsing). But without the bridge ...
If something similar happens to you, it will save a few minutes and your blood pressure.
P.S.:
I must praise their network support that it took only about 2 hours to report the problem, and the profile was changed correctly.
In fact, it is better to make such changes over the weekend. Then, there is plenty of time to find solutions.
Wrinkles on my forehead were caused by the fact that the ISP router was accessible from the WAN. And you know what I think about a toy from ISPs, which they call a router.
What preceded it:
- A few hours before that, I ordered an upgrade of my primary FFTx over the Internet
- 300Mbps upload at the cost of the original 150Mbps
- Two hours later, there was a short WAN outage because the profile was modified at the ISP. I expected it, and I also received an SMS. Provisioning was done.
- after about one minute, everything was running.
What I discovered a few hours later:
- I couldn't access my services over the WAN
- the same services were available via LAN
- step by step Network controller - OK
- Nginx proxy manager, Pi-hole, internal DNS, ... - OK
- primary NAS, where all services are available from LAN - OK
I started to get nervous. And hungry. Two things you shouldn't meet in a man.
Bingo! The last of the tests where I used my public fixed IP address and port 80 found the problem. Surprisingly, I looked at the landing page of the ISP's router.
Reason:
When changing a subscriber's profile, it is crucial to consider the possible settings during the provisioning process of the customer's order (ISP side). ONT BRIDGE mode was to be one of the preserved parameters (when it was used before). Unfortunately, I probably discovered a hole in the ISP's provisioning system. So, yes, the bridge mode hasn't been preserved.
Logically, my own router was in DHCP WAN mode, so everything ran from my LAN in terms of services without the need for NAT (normal browsing). But without the bridge ...
If something similar happens to you, it will save a few minutes and your blood pressure.
P.S.:
I must praise their network support that it took only about 2 hours to report the problem, and the profile was changed correctly.
In fact, it is better to make such changes over the weekend. Then, there is plenty of time to find solutions.
-- post merged: --
Wrinkles on my forehead were caused by the fact that the ISP router was accessible from the WAN. And you know what I think about a toy from ISPs, which they call a router.
