[-] ghoarder@alien.top 1 points 10 months ago

I'm not sure I see what the need for FlareSolverr is, as I understand it it's a reverse proxy that will add the cloudflare authentication cookies so that your *arr apps can access prowlarr, however you should just be able to stick them on a named bridge network and access them internally by container name.

I think using FlareSolverr is adding unnecessary complexity as you shouldn't need to make an internet round trip for containers to access each other.

I have all my internet facing stuff behind Authelia with 2fa but it still works because the containers talk to each other on their own network directly.

[-] ghoarder@alien.top 1 points 10 months ago

I wrote something that can setup caddy automatically from docker labels.

It's not well documented as I mostly wrote it for myself. https://hub.docker.com/r/mheys1/docker-dynamic-caddy https://github.com/mattheys/ddc

It basically acts like a DNS server serving up SRV records that caddy can use for dynamic configuration, I added in an on_demand_tls endpoint as well so that you don't get spammed for non existent TLS records.

2
submitted 10 months ago by ghoarder@alien.top to c/main@selfhosted.forum

Hi All, hope you can help.

I'm trying map out how our people data flows around the company. Where to goes and where it comes from.

We have multiple sources of truth, our hr system will be responsible for things like names, Cisco for phone numbers, exchange for emails etc.

I need to map out all the individual connections, so the HR system might import the email from AD, and something else might take the email from exchange and put in a custom database, then something else might take it from the custom database and put it into AD.

Basically it's a mess, I need to be able to document it so we can effectively rip it out and rewire everything much more efficiently going forward, hoping to move maybe a hub and spoke design with multiple producers/consumers where we can easily plumb something new in, or swap out a product without having cold sweats.

I'd like the tool to be less about actual "drawing" and more about me describing the layout and it producing the image output, that way edits to this mess of spaghetti will hopefully be a lot easier until it's done.

[-] ghoarder@alien.top 1 points 11 months ago

I see no one has mentioned Kong, it's an api gateway that can reverse proxy your api requests but it's so much more than that. It can do all sorts of pipelining to do stuff like add JWT, OAuth or Key authentication, rate limiting plus more. You can run it in a docker container too.

[-] ghoarder@alien.top 1 points 11 months ago

If you are writing your own playwright tests do you have the skills to wrap these up in a web api? If so you can then use uptime-kuma to call your playwright tests and return 200 if it's ok or something else if not, Kuma will show the status code history in the dashboard so you could use specific codes for specific reasons, just try and keep them close to official. Or you could setup NodeRed to log more detailed responses to a database but you would then need to do more work to make a pretty way of displaying that, Grafana maybe?

ghoarder

joined 11 months ago