Elektrine
Log in Register
Paige Chat Timeline Gallery Friends Email Drive DNS Private DNS Domains VPN Kairo Nerve
Remote

Chris Siebenmann

@cks@mastodon.social
mastodon 4.8.0-nightly.2026-10-06
  • Open on mastodon.social

That cks. Overcommitted sysadmin, photographer, bicyclist, and other multitudes. I write a lot of words for a programmer. he/him/they/them 🇨🇦

2647 Followers
364 Following
50 Posts
Joined April 03, 2017
Techblog:
https://utcc.utoronto.ca/~cks/space/blog/
Github:
https://github.com/siebenmann
Flickr:
https://www.flickr.com/photos/22276923@N06/
Open post
Chris Siebenmann @cks@mastodon.social
· 1d ago
Replying to
@joe@f.duriansoftware.com @rk@mastodon.well.com I was going to say it was the history of how things developed¹ but then I remembered another big issue, which is ordering. /etc/ttys starts everything at once, while when real daemons started to appear (especially in BSD) they had startup ordering requirements, which were easier in /etc/rc. ¹ V7 was already running some things out of /etc/rc if I remember right.
2
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3w ago
It feels like I should get a replacement for gnome-terminal. Potential candidate programs must fully support Unicode, full terminal colours, and all of that stuff, and run on X (including remote X, ideally). A standalone process would be ideal. There's nothing wrong with gnome-terminal, it's an okay default for when I'm not using xterm, but I'm increasingly disenchanted with Gnome. (konsole is not an option, I'm even less enthused about how KDE works in my environment.)
3
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 2mo ago
Replying to
@ombialik@mastodon.world I have the impression that part of the Roman obsession comes from later European history, where part of the Renaissance was people looking back to the glories of Rome, especially in Italy, in contrast to their current dark situations¹. Then the Renaissance itself was appropriated by later ages in European history and dragged the Roman connection forward. ¹ https://www.exurbe.com/black-death-covid-and-why-we-keep-telling-the-myth-of-a-renaissance-golden-age-and-bad-middle-ages/ and @adapalmer@wandering.shop 's work generally touches on this a fair bit.
exurbe.com

Black Death, COVID, and Why We Keep Telling the Myth of a Renaissance Golden Age and Bad Middle Ages – Ex Urbe

13
2
4
0
Open post
Chris Siebenmann @cks@mastodon.social
· 2mo ago

Today I learned that SunOS¹ nd is even more cursed than I had vaguely remembered it as. It explicitly uses its own IP protocol instead of UDP so that it doesn't have to checksum the data.

¹ I really mean 'SunOS' here, not Solaris. I believe that Sun dropped nd no later than SunOS 4.

7
3
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
I should say this out loud: a program's configuration files and configuration file format is part of its user interface. Much like other user interfaces, you cannot necessarily use a generic 'UI' for your configuration files without inflicting pain on people operating your software. Yes, this means that sometimes you have to design and build your own configuration file format, much like you may have to build other UIs for your program.
18
2
10
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3mo ago

One thing I'm trying to do more often is to do my casual location lookups on openstreetmap.org instead of Google Maps. Do I want to remind myself where Church & Hayden is in Toronto? I can do that with OSM just fine.

7
3
1
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3mo ago
Replying to
@mattblaze@federate.social I guess it also makes it definite that the reading is from this patient and you haven't accidentally selected printing out a saved trace from the previous patient or something.
6
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
If your program's configuration file format is JSON, you're openly advertising that you care far more about programming convenience in reading and loading your configuration file than you do about the people operating your software. "You can generate our JSON with software from something else", yeah, no. You've told me what your priorities are and I'm going to believe you. I would rather run software that actually cares about the people running it.
14
19
4
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@whitequark@social.treehouse.systems It turns out "stop worrying about TLS certificates (because reliable automation¹) for free" is an extremely compelling pitch. But the eventual tsunami is kind of scary. Thinking about it, I suspect the real issuance growth source was that a ton of people who never would have gotten (proper) TLS certificates for things suddenly started getting them, because free and easy. ¹ some disclaimers apply
9
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@peterb@mathstodon.xyz @gnomon@mastodon.social My theory about what is perhaps my most successful program ever is that it became so successful partly because it let people pretend that they weren't pirating the software that made it unnecessary. (I only found out about its success years and years after the fact. Someday I will write the full story for my techblog.)
8
1
2
0
Open post
Chris Siebenmann @cks@mastodon.social
· 2mo ago
AI/LLM stuff
Today's 'fun' math: graduate students (& others) using LLMs and LLM agents on shared Linux servers + LLM agents that "want" to install software system-wide because that's what READMEs and so on say to do + LLM agents that are increasingly willing to do whatever it takes to achieve a goal + Frequent Linux kernel local privilege escalation vulnerabilities Are we going to get hacked by an agent so it can install some random thing for its slop code development? Maybe.
2
0
1
0
Open post
Chris Siebenmann @cks@mastodon.social
· 2mo ago
Ubuntu: you can totally continue to use GNU Coreutils in 26.04 LTS. Also Ubuntu: build-essential depends on the new Rust coreutils. Yeah, that's not "you can totally continue to use GNU Coreutils", although it sure is tempting to build my own build-essential package with a different dependency.
2
1
5
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@jdm_ @whitequark Us, more than ten years ago: "We need a private TLS certificate authority root certificate for our VPN servers, surely the default lifetime of ten years is good enough." Us, ten years after generating the certificate: "Who could have possibly foreseen that a computer science department would still be here ten years later." (Okay, 'and using the same VPN systems'. Our new CA root certificates have a much longer lifetime, although not 100 years for TLS reasons.)
5
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 2mo ago
Once again I'm being reminded that if you drink enough coffee for long enough, all coffee becomes decaf¹. Or maybe I'm a weird space mutant. There are some other signs.² ¹ assuming i'm not just that tired all the time. ² for example, all forms of alcohol I've ever tried have had a distinct, unique taste that is wretched bleah. this is handy for keeping me from the temptation of drink, but. (hard cider is okay but that's because I love cider, I tolerate the alcohol taste for the rest of it.)
1
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3mo ago
Replying to
At one point in this ride I set what may be a personal record for high pedal cadence in a serious situation, because I forgot to gear up for a mild ascent after a dip. The result left me spinning at over 130 RPM while going at 30+ km/h from the momentum (these two are related). Could I have coasted up (part of) the ascent? Yes, of course, but that's not me, because something in my gut says "it's a hill, if you slow down it's all over, go go go".
2
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3mo ago

The joke is on this plague of Googlebot impersonators trying to pillage my techblog: I specifically restrict Googlebot's User-Agent to its published IP address ranges. Had you used another User-Agent, you might have flown under the radar.

(Not Bing's, though, I restrict it too.)

1
2
1
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@gnomon@mastodon.social I keep wishing Bash had added a "getopts_long" builtin so my shell scripts could support long options. It feels so wrong that none of my shell scripts have them. (AFAIK you really need to do this with a builtin, not an external program, or you wind up in quoting hell.)
2
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@agwa I'm fond of the simpler SCGI, but FastCGI has the mindshare (and the implementations) and I entirely agree with you that it's better than HTTP.
2
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@jef@mastodon.social @untitaker@gts.woodland.cafe I think tab separated values (with comments, I hope) are a perfectly good format; I have a lot of configuration file formats in my own tools that are basically lines of 'a b'. What you turn them into internally is not my (sysadmin) problem.
2
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@fanf@mendeddrum.org It warms my heart that mg is still out there, plugging away. I still have dotfiles for it in my various $HOMEs (which have been there for a long, long time).
2
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3mo ago
In the get people Fedi-hired category, my university department has an opening for a senior (research-side) sysadmin where you'll get to do fun things in a dynamic group that's doing stuff at the crossover of chemistry, robotics, and computing. It's sadly not open to remote people, so Toronto only. And you'll be my indirect colleague (not in my group but working closely with us). https://jobs.utoronto.ca/job/Toronto-Research-Computing-Support-Specialist-ON/603316117/ (boosts welcome)
jobs.utoronto.ca
1
0
8
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@coolandnormal @mayintoronto People in Toronto love flat roofs for houses and I have no idea why. Have they not noticed that it snows, sometimes heavily, and then the snow melts? (I understand flat roofs are supposed to be gently sloped and have all sorts of things to manage water and etc, but ... sloped roofs strike me as much more accident-of-construction-and-etc proof. Achieving a water backup on a sloped roof is going to take real work.)
2
4
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 4mo ago
Replying to
@ceejbot@toot.cat I am more on the side of "LLMs are disastrous, oversold snake oil". They're like old-time cure-alls that were loaded up with fun ingredients like cocaine. It's not like you were getting talcum powder and sometimes the cocaine was very effective, but ... (If LLMs did nothing they'd have probably died by now, the problem is that they actually do produce useful results under the right circumstances. At ruinous ultimate costs and with terrible large scale effects and ...)
1
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@earnestma@tech.lgbt @bpettis@mastodon.benpettis.ninja The front page may age out, I think the specific link for this incident is https://www.systemstatus.utoronto.ca/incidents/c7mr1g9nr6sr But that whole set of pages is more sysadmin-focused and formulaic than normal people focused (although it's discoverable by normal people). The linked to utogether page has a message about "Cybersecurity incident involving Quercus parent company " (but no specific durable URL for just that, it'll presumably age out).
systemstatus.utoronto.ca
1
0
1
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@darkuncle@infosec.exchange @mayintoronto@beige.party @stairjoke@indieweb.social @tg@indieweb.social I'm now tempted to write a finger service that exposes specific information on my own machines. (Local 'finger' tells people too much about too many logins, I'd want it to just advertise fixed information, which I could technically do with a shell script and a directory of pre-canned finger output, now I'm more tempted.)
1
0
1
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@digitaylor One of them was me, but the real credit goes to @Ranjit for https://friend.camp/@ranjit/116466501867438169 which is where I discovered you. I'm looking forward to a full listen.
friend.camp
1
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@mayintoronto@beige.party This is my surprised face that I may not be as underpaid as an academic sysadmin as I expected, although those are explicitly starting salaries. (I could look at private sector sysadmin job listings to be better informed, but that would take work and I'm extremely unlikely to ever try to move.)
1
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@mayintoronto@beige.party Based on extremely cursory searches on what Internet searches found me, I might not even be obviously underpaid. (I assume that anyone whose jobs appear on easily found aggregator sites aren't paying top dollar for top talent, senior sysadmins or otherwise, but even then if I was dramatically underpaid, I'd expect visible jobs to meet or casually exceed my current salary. On the other hand they all want cloud experience and etc and that is not me for (university) reasons.)
1
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@gnomon@mastodon.social Long options are where I start thinking "maybe this shell script can be a Python program instead, how hard is it to run some commands?" (But the last time I came close, I wound up with a shell script version and no long options.)
1
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@agwa@follow.agwa.name Mostly I have Opinions™ about FastCGI's protocol, or more specifically about its excessively "clever" encoding schemes. I wrote my own simple SCGI implementation (for both ends), which would have been a lot more work for FastCGI. For bigger load, SCGI's one connection per request thing is a downside, especially if you're forwarding off the local machine. (And even on the local machine, file descriptors go brrr.)
1
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@davecb@hachyderm.io I think TOML's available nesting gets increasingly unclear as you add more levels of nesting. Once you go three or four or five levels deep in a configuration hierarchy, I think TOML is less than ideal and you want something that makes the nesting structure more explicitly visible in text.
1
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@curious_carrot@mastodon.social YAML is better than JSON but not particularly good. TOML is better than YAML if you have something that it's simple enough for. UCL¹ might work okay for somewhat straightforward configuration files. Otherwise, no, you cannot have a generic configuration file format. You will have to design and parse your own, one that takes into account both general lessons learned and the specific configuration needs of your program. ¹ https://github.com/vstakhov/libucl
GitHub

GitHub - vstakhov/libucl: Universal configuration library parser

Universal configuration library parser. Contribute to vstakhov/libucl development by creating an account on GitHub.

1
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@untitaker@gts.woodland.cafe JSON5 is moderately less hostile than JSON but for most programs it's still advertising that you care a lot more about programming convenience than having a good experience of writing configuration files. (For both simple and complex configuration files there are much better syntaxes that are easier for people to deal with.)
1
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
Oops, I have to retract some of my 'the spammers are showing up on schedule' snark, because our primary MX greylists people sometime and if the primary MX is 4xx'ing things, trying the backup MX is reasonable. (But surprise, the backup MX will greylist you too because our MXes are running the same configuration.)
1
3
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@ibboard@hachyderm.io This is one of the things I love about writing my commit messages in Magit in GNU Emacs; magit shows a second pane with all of the changes I'm committing. I find it fairly twitch-inducing to commit without seeing my diffs outside of simple situations.
1
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@davecb@hachyderm.io I don't like YAML all that much but at least it has comments and forces people to write multi-line things and so on. It's more or less a tolerable universal solvent at this point, although it's not actively good the way a configuration file format could be. (If you don't need nesting there are simpler general formats like TOML.)
0
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@mhoye@cosocial.ca As a sysadmin I have some sympathies for Railway (but it needed better documentation). If what you're protecting against is 'we DROP TABLE'd in the production database', then in-volume snapshots for fast restores are perfectly okay as 'backups'. We use ZFS snapshots for this ourselves (and off-system backups too). (Also, I can imagine customer support calls of "I deleted this volume, why are you still charging me for backups for it, what do you mean I have to delete them separately?")
0
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@vathpela@infosec.exchange Curious minds wonder what sort/make/etc of retroreflective jacket you have. (I have the obvious candidate for a cyclist, or at least what was once the obvious candidate, but maybe there's something better these days.)
0
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@curious_carrot@mastodon.social Configuration files are one of the user interfaces of your program, and just as you can't necessarily use a generic UI, you can't necessarily use a generic configuration file format.
0
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 3mo ago
Replying to
@ripienaar@devco.social I have that on my phone and it's what I reach for there. My GMaps/OSM usage is on my desktops.
0
2
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 2mo ago
Replying to
@BladeWDR@infosec.exchange For a proper syndication feed fetcher¹, the only thing that reduces or eliminates 429s is making feed requests less often. Once an hour is the most it should be doing; any faster risks 429s. ¹ one that does conditional GETs and accepts compressed responses, which Miniflux does as far as I can see from server logs.
0
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@gnomon Infrastructure team: ... have you looked at how many failing tests that have been failing for months that some other groups have? (Now I'm curious what their metrics for 'resources used by failing tests' are telling them. And what such metrics could tell people if they were surfaced generally. Presumably a steady upward trend is bad news ...)
0
5
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@vathpela@infosec.exchange Ooh, I see that they have additional retroreflective stuff these days. Strictly speaking I don't need a reflective frame bag, but, you know ... I'm sure I could find something to do with it. And there's the waist/cross body pack, I already use one of those. Temptations.
0
0
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@gnomon@mastodon.social My first reaction was "my gods, that works?" (Well, for long options. For short options you have other issues. But now I'm wondering about nested getopts and other terrible hacks.)
0
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
I did some more digging using our firewall PF logs and it appears pretty definite that some people showed up to do SMTP authentication probes only after this host appeared in DNS MX and got a TLS certificate. It's possible that the TLS certificate is the trigger for SMTP auth attempts, but they're very bad SMTP auth attempts (they aren't starting TLS, for a start, and this backup MX doesn't do SMTP auth).
0
1
0
0
Open post
Chris Siebenmann @cks@mastodon.social
· 5mo ago
Replying to
@gnomon @mhoye I can actually see a potential use for 'we know this test fails consistently and we're investigating so you don't need to keep re-running it all the time', but even if you want to do that, it should be driven by your team, not the infrastructure team, because they don't know if you need to keep re-verifying the failure (... because maybe the rack PDU got replaced again and suddenly it will start working).
0
2
0
0
Back
313k7r1n3
Elektrine

Tor hidden service

elekhj7afj4qnrr4yd3bkzslsyo5jgfxw3orgjkhlcxifueodybyiiad.onion

I2P eepsite

j6b6cyk6gjmepjih7jjadxgxvvf3lzzujljuu2v4biemzpg3naya.b32.i2p

Platform

  • Email
  • Chat
  • Timeline
  • VPN
  • DNS

Company

  • About
  • Contact
  • FAQ
  • Lite (no JS)

Legal

  • Terms of Service
  • Privacy Policy
  • Transparency Report
  • Report Abuse
  • Warrant Canary
  • VPN Policy

Support

  • support@elektrine.com
  • Report Security Issue
Mail client setup IMAP mail.elektrine.com:993 POP3 mail.elektrine.com:995 SMTP mail.elektrine.com:465
© 2026 Elektrine. All rights reserved. Server: 02:27:20 UTC