Basically, yes. You can configure most cron programs to mail task output to you (it’s usually done by setting the MAILTO variable in the crontab, provided sendmail is available on your system).

I use that to do things like:

0 9 11 10 * echo 'lunch with John Doe at 12:20'

It sends me a mail, and I can see the upcoming events with crontab -l. If it’s not a recurring event, I then delete the rule.

My favorite cost cutting tip is to avoid big webapps running on docker, and instead do with small UNIX utilities (cron instead of a calendar, text files instead of note taking app, rsync instead of a filehosting dropbox-like app, simple static webserver for file sharing, etc). This allows me to run my server on a simple Raspberry Pi, with less than 500mb of used RAM in average, and mininal energy consumption. So, total cost of the setup:

• Raspberry Pi : 77€ x 2 = 144€ (I bought two to have a backup if the first one fails)
• MicroSD 64gb : 13€ x 2 = 26€ (main and backup)
• average energy consumption : 0.41€ (2kWh) per month

With that, I run all services I need on a single machine, and I have a backup plan for recovery of both hardware and software.

Getting used to a UNIX shell and to UNIX philosophy can take some time, but it’s very rewarding in making everything more simple (thus more efficient).

There hasn’t been a new Git repo launch in almost a decade

Am I the only person annoyed they seem to mistake repositories for forges? It’s already annoying when casual users say “git” for “GitHub”, but those guys actually want to build a forge, explaining they’re going to do better than anyone else. Maybe start by properly using the terms?

Thanks for the link!

Interesting. I’ve been using ARM cores for low energy for so long that I haven’t bother to look if things were changing. :) How does it perform on the heat dissipation front? Because if it requires a fan, this should be included in the energy consumed.

I assume it’s a x86 CPU, being an intel processor? How does it manage to be as energy efficient as an ARM CPU? Or maybe it’s only when it’s idle?

I do have to say for the purpose of tinkering I love these bigger projects because you learn so much on the way. Now having read your answer I am even more exited to try it out :D

That’s awesome to hear! Welcome, and have fun! :)

I haven’t heard of most of your abbreviations/term till now

Oh, my apologies. Here is a definition list :

• SMTP : Simple Mail Transfer Protocol : the base of any mail system, it’s the server you contact to send emails, which relays your mail to an other SMTP server (where your contact is hosted), which stores the mail for user to retrieve
• IMAP : Internet Message Access Protocol : one of the protocols that can be used to retrieve emails from your mailserver (the other one being POP3)
• SPF : Sender Policy Framework, a configuration on your domain name specifying which machines are allowed to send mails in its name
• DKIM : DomainKeys Identified Mail : a signing process (signing each mail) to validate the “From” email address is indeed authorized from the domain it pretends to
• DMARC : a warning system to let you know when someone pretended to be you (also giving instructions about what to do with emails when SPF and/or DKIM are missing or wrong)

I guess slapping it on my local raspberry pi wouldn’t be enough no?

Oh no, that would be way not enough. :) Managing a mailserver is a sysadmin task by itself. While you don’t need to do much once it works (which often is a perk of sysadmin work, compensating for the fact that when it does not work, they may have to wake in the middle of the night to fix it), it’s notoriously difficult to get right : you have the configuration of the mailserver to get right first, so that you can send emails, but nobody else can and you don’t become a spam relay without knowing it. Then you have a lot of configuration to do to be able to retrieve your emails from your server, which uses other protocols that you must learn about. Then you have “optional” things that you must setup (SPF, DKIM and DMARC), which you won’t be able to send mails to gmail or outlook if you don’t set them up properly. And when you will have got all of that right, you will have enough experience to be hired as a sysadmin. :)

I can’t provide a good resource for learning it, I learned it 15 years ago when it was way more simple (before SPF and DKIM), and picked every addition as they appeared, but any course on how to manage a mail system will do. There is no difference in doing it for your self-hosted server and for a company (except maybe that for a company, they’ll make you handle users in a database, which you can forego for your own needs). I would recommend to learn how to use postfix first, then any imap server (courier-imap is a top runner), and when you’re comfortable with that, you can learn about SPF, then DKIM, then DMARC. But be aware before going through it that this is basically learning a new skill (sysadmin). You can find docker images that setup everything automatically for you, but I would recommend against that, because at some point, things will break and you will have no idea how to fix them. And if you try to fix them while not knowing well what you’re doing, that’s a good way to end up being a spam relay. Plus, those docker images are difficult to customize, which quite defeats the point of managing your own mail system to begin with.

Well I didn’t want google to read my mails

Sadly, it only works if no one in the recipients of the mail is on gmail (or if everyone use pgp, which I would tend to think is even more rare).

I host my own mailserver as well, and I would add as benefits:

• creating as many email address as you want easily, possibly regexp based address (awesome to give every site a different address and know where the spam comes from, without using the well known schema username+something@host). That also makes routing/filtering mails way more easy, you just have to match the recipient address.
• delivering mails to software, to put email at the center of interapps messaging (basically, that means that postfix pass a matching email to the executable of your choice on your system instead of storing it in your mailbox)
• advanced rules for handling emails. When I want to block a spammer that managed to get my real email, I use regexps to match their mails and reject it with a “REJECT 5.1.1 Recipient address rejected: User unknown in local recipient table” error, imitating the error for unknown users, which often triggers a mail system to remove your address from their database
• easily configure apps to send me email. When I write an application that will send emails to me and only me, I configure it to use my smtp on port 25 without authentication instead of the usual smtps configuration they expect. It connects to it and asks to send a mail to me, which is accepted since I’m a local user. It makes everything way easier (try to do that with gmail and get your IP banned)
• easy backups. Both of the mail system (I backup the whole sdcard of the pi) and of the emails. Never lose an email again.

Oh, I see. Totally makes sense. :)

I guess it depends on the country, but here in France, yes, most landline ISPs provide static IPs (maybe all? there are a couple I haven’t try ; mobile IPs are always dynamic, though). It was not always the case, but I haven’t had a dynamic IP since the 2000’. I feel you, dealing with pointing a domain to a dynamic IP is a PITA.

Ahah, yeah, I protected myself against accidentally banning my own IPs. First, my server is a Pi at home, so I can just plug a keyboard and a screen to it in case of problem. But more importantly, as I do that blacklisting through fail2ban, I just whitelisted my IPs and those of my relatives (it’s the ignore_ip variable in /etc/fail2ban/jail.conf)., so we never get banned even if we trigger fail2ban rules (hopefully, grandma won’t try to bruteforce my ssh!). It allowed me to do an other cool stuff : I made a script ran through cron that parses logs for 404 and checks if they were generated by one of the IPs in that list, mailing me if it’s the case. That way, I’m made aware of legit 404 that I should fix in my applications.

Oh, ok, you whitelist IPs in your firewall. That certainly works, if a bit brutal. :) (then again, I blacklist everyone who is triggering a 404 on my webserver, maybe I’m not the one to speak about brutality :P ) You don’t even need a VPN, then, unless you travel frequently (or your ISP provides dynamic IP, I guess).

I’m not sure about the feasibility of this (my first thought would be that ssh on the host can be accessed directly by IP, unless maybe the VPN software creates its own network interface and sshd binds on it?), but this does not remove the need for frequent updates anyway, as openssh is not the only software that could have bugs : every software that opens a port should be protected as well, and you can’t hide your webserver on port 80 behind a VPN if you want it to be public. And it’s anyway a way more complicated setup than just doing updates weekly. :)

If you do not neglect updates, then by all mean, changing ports does not hurt. :) Sorry if I may have strong reaction on that, but I’ve seen way too many people in the past couple decades counting on such anecdotal measures and not doing the obvious. I’ve seen companies doing that. I’ve seen one changing ports, forcing us to use the company certificate to log in, and then not update their servers in 6 months. I’ve seen sysadmins who considered that rotating servers every year made it useless to update them, but employees should all use Jumpcloud “for security reasons”! Beware, though, mentioning port changing without saying it’s anecdotal and the most important thing is updates, because it will encourage such behaviors. I think the reason is because changing ports sounds cool and smart, while updates just sound boring.

That being said, port scanning is not just about targeted pentesting. You can’t just run nmap on a host anymore, because IDS (intrusion detection systems) will detect it, but nowadays automated pentesting tools do distributed port scanning to bypass them : instead of flooding a host to test all their ports, they test a range of hosts for the same port, then start over with a new port. It’s half-way classic port scanning and the “let’s just test the whole IP range for a single vulnerability” that we more commonly see nowadays. But they are way harder to detect, as they scan smaller sets of hosts, and there can be hours before the same host is tested twice.

The best you can do to know if it was an attack is to inspect the logs when you have time. There are a lot of things that can cause a process going wild without being an attack. Sometimes, even filling the RAM can cause the CPU to appear overloaded (and will freeze the system anyway). One simple way to figure out if it’s an attack : reboot. If it’s a bug, everything will get back to normal. If it’s a DDoS, the problem will reappear up to a few minutes after reboot. If it’s a simple DoS (someone exploiting a bug of a software to overload it), it will reappear or not given if the exploit was automated and recurring, or was just a one-shot.

The fact that both your machines fell at the same time would tend to make think it’s an attack. On the other hand, it may just be a surge of activity on the network with VPSes with way not enough resources to handle it. Or it may even be a noisy neighbor problem (the other people sharing with you the real hardware on which your VPSes run who will orverload it).

However Port 22 should never be open to the outside world.

Wat. How do you connect with ssh, then? You can bind openssh to an other port, but the only thing it changes is that you have less noise in your logs. The real most important security measure is to make sure your softwares are always up to date, as old vulnerable software is the first cause of penetration (and yes, it’s better to deactivate password login to only use ssh keys).

Oh, ok. Thanks for letting me know. 👍️

I’ve been running my own email server for years, and while it’s indeed difficult at first, it is possible and you don’t have much to do to maintain it when it works. All the horror stories you hear come from the fact it’s difficult to get right, and even when you get it right, you will have deliverability problems the first year, until your domain name gets established (and provided you don’t use it for spam, obviously - and yes, marketing is spam).

What you need :

• being willing and serious about reading lot of documentation
• an IP that is not recognized as a home IP. So you’ll need a “business ISP”, or one that is not well known. You bypass this problem by using AWS.
• choosing a well recognized TLD for your domain name, like .com, .org, .net, etc. Don’t use one of those fancy new extensions (.shop, .biz, etc), they are associated with spammers.
• learning how SPF works and getting it right (there are plenty of documentation and test tools for that)
• same for DKIM
• same for DMARC

Start using that for a year without making it your main address. Best is to use it for things not too mainstream, like FOSS mailing lists, discussing with people having their own mailserver, etc, those will not drop your mails randomly. When a year has gone with frequent usage, you can migrate to that email address or domain.

Regarding the architecture of your network : do you read your emails on several machines (like, on mobile and laptop)? If not, you can dramatically simplify your design by using pop3 instead of imap, connecting your client to the AWS server, downloading all your emails to computer and removing them from the server at the same time. There, you have all your mails locally and you don’t need dovecot. :)

I don’t use a pihole, but I have a pi with my favorite distro acting as server, and I use dnsmasq for what you mention. It allows to set the machine as the nameserver for all your machines (just use its IP in your router DNS conf, DHCP will automatically point connected machines to it), and then you can just edit /etc/hosts to add new names, and it will be picked up by the nameserver.

Note that dnsmasq itself does not resolve external names (eg when you want to connect on google.com), so it needs to be configured to relay those requests to an other nameserver. The easy way is to point it to your ISP nameservers or to public nameservers like those from Cloudflare and Google (I would really recommend against letting them know all domains you’re interested in), or you can go the slightly more difficult way as I did, and install an other nameserver (like bind9) that runs locally. Gladly, dnsmasq allowed to configure its relay nameserver to be on something else than port 53, which is quite rare in dns world. Of course, if you’re familiar with bind9, you could just declare new zones in it. I just find it (slightly 😂) more pleasant to work with /etc/hosts.

It’s coming to Gitlab too! (although, this will take quite some time)

Obligatory check : are you sure you really need a forge? (that’s the name we use to designate tools like Github/Gitlab/Gitea/etc). You can do a lot with git alone : you can host repositories on your server, clone them through ssh (or even http with git http-backend, although it requires a bit of setup), push, pull, create branches, create notes, etc. And the best of it : you can even have CI/CD scripts as post-receive hooks that will run your tests, deploy your app, or reject the changes if something is not right.

The only thing you have to do is to create the repos on your server with the --bare flag, as in git init --bare, this will create a repos that is basically only what you usually have in the .git directory, and will avoid having errors because you pushed to a branch that is not the currently one checked. It will also keep the repos clean, without artifacts (provided you run your build tasks elsewhere, obviously), so it will make all your sources really easy to backup.

And to discuss issues and changes, there is always email. :) There is also this, a code review tool that just pop up on HN.

And it works with Github! :) Just add a git remote to Github, and you can push to it or fetch from it. You can even setup hooks to sync with it. I publish my FOSS projects both on Github and Gitlab, and the only thing I do to propagate changes is to push to my local bare repos that I use for easy backups, they each have a post-update hook which propagates the change everywhere it needs to be (on Github, Gitlab, various machines in my local network, which then have their own post-update hooks to deploy the app/lib). The final touch to that : having this ~/git/ directory that contains all my bare repos (which are only a few hundred MB so fit perfectly in my backups) allowed me to create a git_grep_all script to do code search in all my repos at once (who needs elasticsearch anyway :D ) :

#!/usr/bin/env bash
# grep recursively bare repos

INITIAL_DIR=$(pwd)
for dir in $(find . -name HEAD -exec dirname '{}' \;); do
  pushd $dir > /dev/null
  git grep "$*" HEAD > /dev/null
  if [[ "$?" = "0" ]]; then
    pwd
    git grep "$*" HEAD
    echo
  fi

  popd > /dev/null
done

(note that it uses pushd and popd, which are bash builtins, other shells should use other ways to change directories)

The reason why you may still want a forge is if you have non tech people who should be able to work on issues/epics/documentation/etc.