g_host

That's a good call out but it is something I tested originally and ruled out.

I've got some applications used daily by friends all over the world and the services would all become unavailable to them when this started happening. Only fix I found was restarting NGINX service and then it could happen again an hour later or 3 days later. Once I removed the proxy config for Hister from the service, the issue never happened again.

I'll repro the issue and get the details intoa. Github Issue this weekend.

Yes, I should have some time this weekend to reproduce the problem and provide useful information for ya'll to look at...even if it's just to determine the problem is for me to fix and not a bug with Hister.

I love Hister so much.

I ran it for most of this year but encountered some problems with it I couldn't fix and thus have not had it hooked up to anything since June when I finally couldn't take it anymore.

In short, I serve a good number of apps from an Nginx reverse proxy. Maybe 25% of them are exposed to the WWW while everything else is limited to the LAN but I still get valid TLS for all of it.

Hister, though, kept breaking my whole reverse proxy and I could never figure out EXACTLY why so I could fix it. After running fine for a few days, it would hog the whole server and everything else proxied by Nginx would become unreachable. I tried tuning the config for it to no avail.

One day when I'm less lazy, I'll probably hook it back up via it's LAN IP to every machine I've got again. I REALLY liked that I could log my browsing history from any machine anywhere in the world without a VPN and I was really disappointed when I had to disable its config in Nginx.

I still use it a lot to go find stuff I flagged as important quickly.

I'm curious if this is something you've heard of before, or if I've got a one off problem here.

I even ported the config to a brand new VM with NGINX and still had the same problem.

I don't understand why Apple won't support Aura Cast already.

"The company formalized the idea that defects could be fixed through human intervention on live production systems" (From Part 5).

Uh...yeah. I think we all realized that years ago.

"The company formalized the idea that defects could be fixed through human intervention on live production systems"

Uh...yeah. I think we all realized that years ago.

Reader mode for the win

It is understandable that one would be frustrated with a Raspberry Pi handling critical network services like DNS.

I've been running pi-hole in KVM guest virtual machines for more years than I can remember and never had any problems. I would expect a Raspberry Pi to eventually choke on the demand of providing 24/7 service to a network.

But not everyone has a hypervisor in their basement. Forking over $20/year is definitely better on the budget than buying a server.

But if you already have a server or some reliable hardware in your LAN, there's no good reason to leave anything important up to a Raspberry Pi.