A while ago I wrote an entry about (not) finding which NFSv4client owns a lock on a Linux NFS(v4) server ,where the best I could do was pick awkwardly through the raw NFSv4 client information in /proc/fs/nfsd /clients. RecentlyI discovered an alternative to doing this by hand, which is the nfsdclnts program, and as a result of digging into it and what I was seeingwhen I tried it out, I now believe I have a better understanding

Over on the Fediverse, I said something :

Thesis: DNS is not meaningfully a database, because it's explicitlydesigned and used today so that it gives different answers todifferent people. Is it implemented with databases? Sure. But treatingit as a database is a mistake. It's a query oracle, and as a queryoracle it's not trustworthy in the way that you would normally trusta database to be, for example, consistent between different peoplequerying it.

It would be nice if

In a comment on my entry about how Unix files have at least twosizes , Leah Neukirchen said that 'ZFS compresses asynchronously'and noted that this could cause the reported block size of ajust-written file to change over time. This way of describing ZFS'sbehavior made me twitch and it took me a bit of thinking to realizewhy. What ZFS does is delayed compression (which is asynchronouswith your user level write() calls), but not true 'asynchronouscompression' that happens later at

I'll start by presenting things in illustrated form:

; ls -l testfile-rw-r--r-- 1 cks 262144 Apr 13 22:03 testfile; ls -s testfile1 testfile; ls -slh testfile512 -rw-r--r-- 1 cks 256K Apr 13 22:03 testfile

The two well known sizes that Unix files have are the logical 'size'in bytes and what

The news of the time interval is that the maximum validity periodfor TLS certificates will be lowered to 47 days by March 2029 , unless theCA/Browser Forum changes its mind (or isforced to) before then. The details are discussed in SC-081 . In skimming themailing list thread on the votes ,a number of organizations that voted to abstain seem unenthused (anduncertain that it can actually be implemented), so this may not cometo pass, especially on the

These days, Unixes are quite generous in what they make availablethrough their packaging systems, so you can often get everythingyou want through packages that someone else worries about building,updating, managing, and so on. However, not everything is availablethat way; sometimes I want something that isn't packaged, andsometimes (especially on 'long term support' distributions) I wantsomething that's more recent that the system provides (for example,Ubuntu 22.04 only has Emacs 27

Back in the old days, it used to be reasonably routine to have'multi-architecture' Unix environments with shared files (wherehere architecture was a combination of the process architecture andthe Unix variant). The multi-architecture days have faded out , and with them fading, so has informationabout how people made this work with things like local binaries.

In the modern era of large local disks and build farms, the defaultapproach is probably to simply build complete copies of '/local'for each architecture

I've written entries criticizing things that support using OIDC (OAuth2)authentication for not supporting it with general OIDC identityproviders ('OPs' in OIDC jargon), only with specific (large) oneslike Microsoft and Google (and often Github in tech-focused things).For example, there are almost no mail clients that support usingyour own IdP , and it's much easier tofind web-based projects that support the usual few big OIDC providersand not your own OIDC OP. However

Over the history of a given Fedora version, Fedora will often releasemultiple updates to the same package (for example, kernels, butthere are many others). When it does this, the older package windup being removed from the updates repository and are no longerreadily available through mechanisms like 'dnf list --showduplicates'. For a long time I used dnf's 'local' plugin to maintaina local archive of all packages I'd updated, so I could easilyrevert,