search

Akkadius / glass-isc-dhcp

Glass - ISC DHCP Server Interface
MIT License
696 stars 144 forks source link

Is project discontinued? #107

Open f18m opened 5 months ago

f18m commented 5 months ago

Hi @Akkadius , I think this project is really nice and I was about to deploy it. However I see that the last commit is 5yrs old. I guess there will be a large number of dependencies that should be bumped to e.g. fix security vulnerabilities.

Are you considering to work on this project? Would you accept collaborators to work on it (honestly my Javascript is a bit rusty but I might try). Also there's a queue of PRs pending I see...

Thanks!!

vrisk commented 5 months ago

ISC DHCP is itself no longer maintained, so that may be a factor in the developer of this tool moving on to other work. ref: https://www.isc.org/dhcp/

f18m commented 5 months ago

thanks @vrisk for the answer. I believe however that ISC DHCP is still the de-facto DHCP implementation. The replacement, Kea, is actually a software solution with a much larger footprint that is not suitable for small deployments (e.g. at your home). I think it's a pity that such project is not maintained. I know it's easy (I do that on most of my open source projects) to setup a Renovate bot and get PRs automatically created at least to maintain up to date the dependency chain.

I wonder if the author @Akkadius would like to get any help in that sense...

Akkadius commented 3 months ago

It's possible I revisit this project at some point.

Originally, I pushed up something thinking that it was useful to my ISP and that it might be useful to others. Clearly, after 7 years it has had quite the uptake. It wasn't long after I published this I left the ISP and moved into big tech.

If there was community support to see this project go further, there's a lot I could do to improve it drastically. I'd end up porting the project to Go so it could be compiled into a portable cross-arch / platform binary. It'd be a far cleaner and easier to maintain project overall in the end. I used this project to learn Node and today its just not my first choice for building servers 7 years later.

I'd build in standard docker deployments along with CI integration testing.

My time is limited and I have a family to support so would have to crowdsource funding to motivate me to do it all - but curious if the desire is there.

Curious others thoughts here - also long time no talk @vrisk !

f18m commented 3 months ago

hi @Akkadius ,

It's possible I revisit this project at some point.

that's nice to hear!

If there was community support to see this project go further, there's a lot I could do to improve it drastically. I'd end up porting the project to Go so it could be compiled into a portable cross-arch / platform binary. It'd be a far cleaner and easier to maintain project overall in the end. I used this project to learn Node and today its just not my first choice for building servers 7 years later.

well Go and docker would be my first choice as well. OTOH I'm not sure if re-designing and recreating from scratch is really worth it...

My time is limited and I have a family to support so would have to crowdsource funding to motivate me to do it all - but curious if the desire is there.

Same here -- I was interested in deploying this in my home automation systems, so there's no business and no money behind that. OTOH I might be able to help as I like to contribute to open-source projects in my spare time... it might be just a few hours per week but again it's all about collaborating and joining efforts...

Akkadius commented 3 months ago

well Go and docker would be my first choice as well. OTOH I'm not sure if re-designing and recreating from scratch is really worth it...

I pumped Glass out over a weekend initially. The port wouldn't take that long. There are more long term risks, maintainability risks of having the project in Node than to just invest in quickly porting it.

With only having so little time invested in the project, I'm amazed it picked up the level of popularity it did.