OS

After my attempts to make a binhost resulted in a overcomplex mess that myself could not maintain, time for a more simplified approach: just let the hosts use emerge as usual, but delegate the heavy lifting to a more powerful machine. Enter distcc. It might be a bit slower than a perfectly tuned binhost, but it is much easier to set up. In this guide, we will walk through setting up a Gentoo client and server using OpenRC and IPv6. Please, for the love of all saints and martys, learn IPv6, let IPv4 die in peace. ...

In the previous guide, we built a dedicated binhost to compile your stuff. Time to setup your servers to consume these packages. In this follow-up guide, we will configure your servers. By pointing their Portage installations to your binhost, these machines will be able to install and update packages in minutes rather than a whole day. Prerequisites & Assumptions Shared Storage: We assume each server has access to a remote folder (via NFS, SMB, or a mounted cloud drive) at a specific mount point we will refer to as <path>. Another layer of complexity by setting in maintaining a side channel for distribution, like rsync thru ssh, is a hassle. Folder Structure: On that path, the binhost has populated <path>/<host>/binpkgs with compiled packages from the last guide and <path>/gentoo with the Portage repository tree. No Local Syncs: Because the repository is shared, these client machines are no longer supposed to run emerge --sync. They will always read the repository state managed by the binhost. The Public Key: Ensure the binhost-public.asc file (generated in Part 1) has been copied to your client server’s current working directory. All that needs to be done only once. ...

Maintaining a fleet of Gentoo systems, or struggling to compile packages on an underpowered server, can be an incredibly tedious process. Waiting hours for a single package to compile can quickly add up, and eventually you may try to burn down your home lab. Unless you rich, your home servers are likely stuff of 10 years ago made of mismatch stuff lying around. A partial solution is simple do not touching the portage flags at all and use the Gentoo provided binhost. But this presents two issues: not everything available on portage exist as binary package at their binhost mirrors, and the second is that you give away customization of the packages that you do use. ...

If you arrived here, it’s because you are tired of script kiddies attempting to brute force your server with sshd and polluting its logs. The good news is that NetBSD has preinstalled a solution to deal with such situations that is more convenient than the famous fail2ban: the blocklistd. The blocklistd is a deamon that has only one life purpose: monitor the number of failed connection attempt with some system service. After a defined threshold of failures, it will block the connection for a predefined time, for example, 6 hours. ...

Supposing you want to host a static pages site or your blog generated with Hugo on bozotic, but need a certificate for HTTPS, given that HTTP is almost dead and buried, here a tutorial on how to do it on NetBSD. First, install acmesh with pkgin: pkgin install acmesh After that, the command acme.sh will be available. Do all of the following as root (su root). Subscribe to Let’s Encrypt using the client, only a email is necessary: ...

Who deals with embedded Linux already had the situation where there’s the need to inspect the contents of a system that executes from RAM. Usually, it exists only compressed as a single file at a boot folder, or is thrown around at an EFI partition or /boot formatted using FAS16/32, like on a RaspberryPi. The command for that is simple, with the initramfs on this example being called rootfs.cpio.zst: zstdcat rootfs.cpio.zst | cpio -idmv That’s it. Since it isn’t really a file system, there’s no need to mount it. It will expand all their root tree contents inside the folder that you called the command, ready to be inspected.

If do you use applications like mdnds on NetBSD on a simple OrangePi Zero, soon will take notice that it will fill the RAM, even if not used for anything. One solution is to activate the good and old swap, who is disabled by default on a OrangePi Zero NetBSD image. As root, do the following commands to create a swapfile with 2Gb. Febronia# dd if=/dev/zero of=/swapfile bs=1m count=2048 Febronia# chmod 600 /swapfile Febronia# swapctl -a -p 0 /swapfile Febronia# echo "/swapfile none swap sw,priority=0 0 0" >> /etc/fstab And that’s it. I should already show on “top” and on /proc/meminfo.

Do you have a dynamic IP connection and your domain service has DDNS? Possible, it also accept DDClient to update your A or AAAA records with your dynamic IP. Unfortunately, of the time of this article, the DDClint at NetBSD repositories only accept IPv4. Thus if you need IPv6, you got a problem. Below, there’s the procedure to install the current version from GitHub on a NetBSD system. First, supposing that you still don’t have installed pkgin to make use of binary packages on NetBSD beyond pkgsrc, who is a overkill just to install a single small application, configure it: ...

Do you know that NetBSD has a simple web server pre-installed called Bozotic? It is very simple, so simple that it can be configured with just a single command line, and quite useful to serve file lists or static web pages. Like this blog. Supposing that you has a valid domain certificate, be using let’s encrypt of something else, who creating and validating is beyond the scope of this tutorial, all that needs to be done is: ...

One of the most irrational things on Linux, when they don’t have to deal with Windows, it why use SAMBA/CIFS. It’s cumbersome to configure, and don’t feel at home on Linux at all. And there’s NFS, that unfortunately ins’t much better at the simplicity aspect (if you use authentication), and actually, can be even worse if you use some exotic distribution without the required kernel modules. But there’s something that all slightly more advanced users has on his file server that usually is shipped by default: SSH. ...