RE: https://infosec.exchange/@david_chisnall/116543566441127430
I cannot begin to count how many words I’ve written about SMTP (and email in general) being unfit for various purposes.
The only sysadmin subspecialty I am notably advanced in is email administration.
I wish I understand *why* what I did fixed the problem that I still don’t understand, but it did.
I was just trying to get all 5 VMs involved onto one host so I could snoop on the traffic between them more easily.
I’m blaming stuck ARP or Spanning Tree. Two of those VMs were briefly fighting over the same IP last night before I fixed the freshly-minted clone to use its proper new one.
2 new chassis, 4 nodes each.
1 stalls in POST.
2 are still untouched (booting to UEFI shell)
2 have been installed but won’t boot from the disks
2 booted after install, but I oopsed something
1 is exactly what I want.
I’d kill for a way to copy BIOS configs & to tell the AlmaLinux installer to do a 3-way mirror on 3 SSDs & slice it according the selected security profile. I guess I need to learn Kickstart...
(I wouldn’t kill a human but I’d kill *something.*)
It’s really appalling how much time I waste trying to figure out which antique browsers on which platforms I need to run to talk to all of the antique hardware that I support.
This message brought to you by PDUs with no solution...
I started out trying to fix a problem between my Nextcloud Desktop client and one server, ended up submitting a patch to MacPorts for an update to the cadaver port. Perfectly normal sequence.
(Nextcloud client DOES NOT use cadaver)
WHY in 2026 am I having to flip switches in Setup to a "legacy" boot to get it to boot from anything but the network???
OK, so these aren’t exactly new nodes. Supermicro X11 boards with Nutanix firmware(?) that are designed to drop into a cloud infra setup. But we don’t have that.
The boot order SAYS try disks of all flavors but it just doesn't do that.
I blame IBM. This whole architecture was a botch from the start and has just gotten slimier as we've "fixed" it.
Just ran out of brainpower for the day trying to fix WTF a dev w/root on a production box did to deploy a #uwsgi application. Dev in question: no longer available. Apparently he didn't know there was a EL package available or how the “Emperor" rig works, so the app needs to be started by hand on every reboot. This was not a documented fact.
I failed to hack up a fix. The Emperor kept cursing the application instance. I like the absurd terminology, hate the documentation...
I don’t care who says it happened, it’s LIBEL!
@spamhaus has never had a “list everything” event by choice. I have used their services since the beginning and I would have noticed.
No, I’m not going to point at the latest instance of that defamation. But wherever you see someone claiming that you can’t trust Spamhaus because of them having done that, know that you’re seeing the words of an idiot or a butthurt spammer.
Oh. Huh. Imagine that….
The .org root DNSSEC is gerbusterficated or in transition or something.
dnsviz is being cranky for me. Everyone else stop using it for the day, OK?
🆘Bill Cole 🇺🇦