It obviously depends on your exact git workflow, but my last team had things setup so that the code content of a MR was automatically squashed on merge, and the text if the MR itself was automatically set as the content of the new singular git commit.

This was largely the best of both worlds because your commits could have almost any text, and the description of what changed could be updated as needed when making the MR. But it ultimately ended up in the git history where it belonged.

Of course, I still had some trouble trying to get the team to describe their changes well in the MR at times - but that’s a different problem entirely.

Bleh, maybe I’m an old man, but when I’m searching stackoverflow, I find the context of stack overflow answers really helpful.

I.E. the top result may include caveats itself or have comments indicating why an answer might be problematic. And sometimes the best answer isn’t even the top answer. I’ve not used AI code assistance very much, but these all seem like things that the model is likely to take for granted.

But I also never contribute to stackoverflow, and agree I’d much rather engage with with an AI than do THAT.

Another solution to this situation is to squash your changes in place so that your branch is just 1 commit, and then do the rebase against your master branch or equivalent.

Works great if you’re willing to lose the commit history on your branch, which obviously isn’t always the case.

Found a blog post that gives a quick overview of how to do git via email in general: https://peter.eisentraut.org/blog/2023/05/09/how-to-submit-a-patch-by-email-2023-edition

So at least from my understanding you’d make your changes, email the contents of the patch to the maintainer, and then they’d apply it on their side, do code review, email you comments, etc. until it was in an acceptable state.

There’s also the full kernel development wiki that goes into all the specifics: https://www.kernel.org/doc/html/v4.16/process/howto.html

(I never got through the whole thing)

I’ll also throw out: aging infrastructure, build systems, coding practices, etc.

I looked into contributing to the kernel - it’s already an uphill battle to understand such a large, complex piece of software written almost entirely in C - but then you also need to subscribe to busy mailing lists and contribute code via email, something I’ve never done at 30 and I’m betting most of the younger generation doesn’t even know is possible. I know it “works” but I’m really doubting it’s the most efficient way to be doing things in 2024 - there’s a reason so many infrastructure tools have been developed over the years.

The barriers to entry for a lot of projects is way too high, and IMO a lot of existing “grey” maintainers, somewhat understandably, have no interest in changing their processes after so much time. But if you make it too hard to contribute, no one will bother.

Out of curiosity, what switch are you using for your setup?

Last time I looked, I struggled to find any brand of “home tier” router / switch that supported things like configuring vlans, etc.

Maybe I am not thinking of the access control capability of VLANs correctly (I am thinking in terms of port based iptables: port X has only incoming+established and no outgoing for example).

I think of it like this: grouping several physical switch ports together into a private network, effectively like each group of ports is it’s own isolated switch. I assume there are routers which allows you to assign vlans to different Wi-Fi access points as well, so it doesn’t need to be literally physical.

Obviously the benefits of vlans over something actually physical is that you can have as many as you like, and there are ways to trunk the data if one client needs access to multiple vlans at once.

In your setup, you may or may not benefit, organizationally. Obviously other commenters have pointed out some of the security benefits. If you were using vlans I think you’d have at a minimum a private and public vlan, separating out the items that don’t need Internet access from the Internet at all. Your server would probably need access to both vlans in that scenario. But certainly as you say, you can probably accomplish a lot of this without vlans, if you can aggressively setup your firewall rules. The benefit of vlans is you would only really need to setup firewall rules on whatever vlan(s) have Internet access.

I ran into the same thing. I’ve always just worked around it, but I believe I did find the solution at one point (can’t find the link now).

But if I am remembering right, I believe you need to manually create a bridge between the two networks - by default it isolates the VMs from TrueNAS itself for security reasons.

Sorry I can’t link the exact fix right now, but hopefully this will help you Google the post I found on the subject.