For reasons outside the scope of this entry , I was recentlyreminded of some very old entries here where I compared somePython code with some Perl code to do the samething . One of the things that stoodout to me is that way back then I said:

For example, I could have written ' print "\n".join(rr.strings) ' inPython, but it doesn't feel right; I would rather write the for loopexplicitly instead.

At some point between back then

I recently wrote about how getting out of being people's secondaryauthoritative DNS server is hard .In the process of that I said that there was a background Internetradiation of external machines throwing random DNS queries at us.Now that we've reduced the number of DNS zones that were improperlystill pointing at us for historical reasons, I think I can finallysee enough from our public authoritative DNS server's traffic to say something about that.

The rejected DNS queries we're seeing so far are a mixture

In response to yesterday's entry on using systemd (service) unitsfor easy capturing of log output , acommentator drew my attention to systemd-run and systemd-cat . Ispent a bit of time poking at both of them and so I've wound upwith some things to remember and some opinions.

(The short summary is that you probably want to use systemd-run witha specific unit name that you pick.)

Systemd-cat isvery roughly the systemd equivalent of logger . As you

Suppose, not hypothetically ,that you have a third party tool that you need to run periodically.This tool prints things to standard output (or standard error) thatare potentially useful to capture somehow. You want this capturedoutput to be associated with the program (or your general systemfor running the program) and timestamped, and it would be handy ifthe log output wound up in all of the usual places in your systemsfor output. Unix has traditionally had some solutions for this, suchas logger for sending

Without deliberately planning it, we've wound up running an assortmentof DNS server software on an assortment of DNS servers. A lot ofthis involves history, so I might as well tell the story of thathistory in the process. This starts with our three sets of DNSservers: our internal DNS master (with a duplicate) that holds both the internal and external views of our zones ,our resolving DNS servers (which use our internal zones), and ourpublic authoritative DNS server (carrying our external zones

Many, many years ago, my department operated one of the university's secondary authoritative DNS servers,which was used by most everyone with a university subdomain and asa result was listed as one of their DNS NS records. This DNs serverwas also the authoritative DNS server for our own domains, becausethis was in the era where servers were expensive and it made perfectsense to do this. At the time, departments who wanted a subdomainpretty much needed to have a Unix system administrator and probablyrun their own

A while back I wrote about how in POSIX you could theoreticallyuse inode (number) zero . Not allUnixes consider inode zero to be valid; prominently, OpenBSD's getdents(2) doesn't returnvalid entries with an inode number of 0, and by extension, OpenBSD'sfilesystems won't have anything that uses inode zero. However, Linuxis a different beast.

Recently, I saw a Go commit message with the interesting description of:

os: allow direntries to have zero inodes on

The X Window System has a long standing concept called 'visuals'; to simplify, an Xvisual determines how to determine the colors of your pixels. As Iwrote about a number of years ago, these days X11 mostly uses'TrueColor' visuals , which directly supply8-bit values for red, green, and blue ('24-bit color'). However X11has a number of visual types , such asthe straightforward PseudoColor indirect colormap (where everypixel value is

DMARC has a feature whereyou can request that other mail systems send you aggregate reports about theDMARC results that they observed for email claiming to be from you.If you're a large institution with asprawling, complex, multi-party mail environment and you're consideringtrying to make your DMARC policy stricter, it's very useful to getas many DMARC reports from as many people as possible. Especially,'you' (in a broad sense) probably want to get as much informationfrom mail