Post AX7uoiI9cPz5FoI71s by mairin@fosstodon.org
(DIR) More posts by mairin@fosstodon.org
(DIR) Post #AX7ubqafSAdTE8od60 by mairin@fosstodon.org
2023-06-26T19:37:32Z
0 likes, 0 repeats
RT from Red Hat:We will always send our code upstream and abide by the open source licenses our products use, which includes the GPL. Red Hat's Mike McGrath expands upon the recent CentOS Stream changes: https://bit.ly/44jIZWc.
(DIR) Post #AX7ubrR8J6bRqrMWsC by aeva@mastodon.gamedev.place
2023-06-27T03:14:46Z
0 likes, 0 repeats
@mairin so I have this right: there's no license violations, all changes are contributed to upstream projects, all the source is also publicly available for anyone who is willing to dig through the git history, and redhat is still bankrolling like half of the development of the mainstream linux ecosystem...?
(DIR) Post #AX7ubsPkfj5msriwMK by resuna@ohai.social
2023-06-27T12:03:39Z
0 likes, 0 repeats
@aeva @mairin Red Hat has "contributed" some stuff that has negative value, like Systemd.
(DIR) Post #AX7ubtEnbvvRRBbhvU by mairin@fosstodon.org
2023-06-27T13:04:41Z
0 likes, 0 repeats
@resuna @aeva OH ARE WE GONNA BRING UP THAT GREATEST HITS OF THE INTERNET NOW??
(DIR) Post #AX7ubu8oFgjEEtoREG by finder@fosstodon.org
2023-06-27T13:27:51Z
0 likes, 0 repeats
@mairin @resuna @aeva Oooehhehehoho og noooo hecking systemd is the deviiiiiiiil nooooieofiieieoeoekoririfioe
(DIR) Post #AX7ubv5IkDW5AJB9Oq by resuna@ohai.social
2023-06-27T13:40:46Z
0 likes, 0 repeats
@finder @mairin @aeva Also TERFS are bad mmmmkay.
(DIR) Post #AX7ubvgsUVZf2ql9yC by mairin@fosstodon.org
2023-06-27T14:37:45Z
0 likes, 0 repeats
@resuna @finder @aeva What does systemd have to do with terfs
(DIR) Post #AX7ubwNlv1sxBspPpQ by aeva@mastodon.gamedev.place
2023-06-27T19:22:50Z
1 likes, 0 repeats
@mairin @resuna @finder well see "Lennart Poettering" sounds kinda like "Harry Potter", and J.K. Rowling is a TERF, so therefore systemd is also a terf. We should all throw our fast boot times into the garbage and go back to that System V thing we used to use, which was better because it gave you more time to meditate on the sagacious words of Richard "Epstein Did Nothing Wrong" Stallman, our glorious leader who has definitely never done anything problematic to women in the community. đ
(DIR) Post #AX7uhsA0WrXBplDcSu by jbowen@mast.hpc.social
2023-06-26T20:10:20Z
0 likes, 0 repeats
@mairin Wow. Really doubling down on their position.
(DIR) Post #AX7uhsp83yQZtISSYq by rijnhard@fosstodon.org
2023-06-26T21:53:39Z
0 likes, 0 repeats
@jbowen @mairin I mean, he has a point. I don't like it, my company can't afford the sheer number of RHEL licences we would need for our server count vs our revenue, that's why we used centos, and it's been 10 years of building on centos as well, so changing isn't easy. I also imagine we aren't the only people out there in that situation, especially post covid. But even with all that, I find it hard to disagree, as much as this pisses me off, he is right in that we need red hat...
(DIR) Post #AX7uhtSpgMBdsR2ARk by jbowen@mast.hpc.social
2023-06-26T22:29:15Z
1 likes, 0 repeats
@rijnhard I'm going to push back strongly on "we need Red Hat[.]"There's a lot of things the free/open community need, but a company acting in bad faith by disrupting things in two consecutive release cycles with zero warning or empathy is not one of them.I'm currently typing this on a Fedora machine. I have commits in software that RH packages and sends to their customers. I will *not* be running Fedora on any of my machines by the end of the weekend.1/2
(DIR) Post #AX7uhucnMnkZTchdc8 by jbowen@mast.hpc.social
2023-06-26T22:30:03Z
0 likes, 0 repeats
@rijnhard I will continue to contribute to projects (on my own free time) on a different OS which RH will package and send out and I won't call them a leech.Never underestimate the vindictiveness of a nerd scorned.2/2
(DIR) Post #AX7uoiI9cPz5FoI71s by mairin@fosstodon.org
2023-06-27T02:10:34Z
0 likes, 0 repeats
@jbowen @rijnhard Can you help me understand how you feel this is in bad faith?These execs just laid off close to 1000 people. To me, as an employee at any comoany, it would be bad faith to continue enabling something that doesn't benefit us and actually costs us, and then layoff more people, instead of tighten belts and do everything possible to avoid laying anyone off.
(DIR) Post #AX7uoj1ssOZ1XdgdJA by mairin@fosstodon.org
2023-06-27T02:11:57Z
0 likes, 0 repeats
@jbowen @rijnhard If Red Hat lays off open source contributors, in *this* economy, what are the chances they get another gig producing open source software? I mean, really? So if Red Hat lays off open source contributors, it impacts everyone relying on that software.
(DIR) Post #AX7uojhMOBjzcH5kxM by jbowen@mast.hpc.social
2023-06-27T02:24:31Z
1 likes, 0 repeats
@mairin @rijnhard I don't know what the chances are. Maybe some things suffer for a bit. But the FLOSS world is resilient and nothing will come crumbling down.Does RH think it's the only one who employs developers who work on FLOSS projects?
(DIR) Post #AX7usxvbQQoEuJlpWy by jbowen@mast.hpc.social
2023-06-27T02:21:32Z
1 likes, 0 repeats
@mairin You're privy to more than I am, but from the outside when Red Hat began sponsoring CentOS in 2014 it looked like an endorsement of that project. Other clones stopped. Then in 2019, shortly into RHEL 8's life cycle, RH cut short the planned life of CentOS 8. Then just as Alma and Rocky are starting to feel really stable, access to git.c.o is pulled. So it seemed like RH was saying "We like CentOS. Sike! But you can still have sources. Sike!"@rijnhard
(DIR) Post #AX7usz8OwKdoeIlZ7Q by jbowen@mast.hpc.social
2023-06-27T02:22:09Z
0 likes, 0 repeats
@mairin @rijnhard Maybe it's just really poor communication rather than bad faith, but from the outside its difficult to tell the difference and people will assume the worst.
(DIR) Post #AX8UK8UnjusjtfvS1g by tml@urbanists.social
2023-06-27T15:31:23Z
0 likes, 0 repeats
@resuna @aeva @mairin OMG, is #systemd hating still a thingâŚ
(DIR) Post #AX8UK9ZnioTXFTGxSS by eliasp@mastodon.social
2023-06-27T15:47:10Z
0 likes, 0 repeats
@tml some people will never get over it, that their collection of redundant and inconsistent and imperative shell-script hacks has been replaced by a set of coherent and declarative units, controlled by a central logic-instance with clear interfaces. đ¤ˇââď¸@resuna @aeva @mairin
(DIR) Post #AX8UKAUWJvqU5NoFrk by resuna@ohai.social
2023-06-27T15:54:47Z
0 likes, 0 repeats
@eliasp @tml @aeva @mairin Absolutely, having services start in an unpredictable order and having VMs randomly hang on startup due to the load on the underlying host is an absolute win.
(DIR) Post #AX8UKBCTgV0WHiNMNk by jorge@hachyderm.io
2023-06-27T22:42:13Z
0 likes, 0 repeats
@resuna @eliasp @tml @aeva @mairin You might want to check the documentation because if systemd actually behaved that way the modern world would have collapsed a long time ago.
(DIR) Post #AX8UKBwutq9cbk6RlY by resuna@ohai.social
2023-06-27T22:48:09Z
0 likes, 0 repeats
@jorge @eliasp @tml @aeva @mairin Iâm sure it wasnât intended behaviour. When we updated RHEL to a version with systemd on a VM cluster we started getting boot failures, and we could see the order of system startup changing from boot to boot. So subsystems were getting in deadlock because the dependencies werenât being satisfied⌠occasionally. We had to assign someone to checking our clusters and rebooting stuff.
(DIR) Post #AX8UKCaGXXd6ZmVs6C by justinz@social.opendesktop.org
2023-06-28T02:19:30Z
0 likes, 0 repeats
@resuna @jorge @eliasp @tml @aeva @mairin this is why VM management tools have delayed startup options, to stagger the load.
(DIR) Post #AX8UKDqbqGIUUlARDE by resuna@ohai.social
2023-06-27T22:52:44Z
0 likes, 0 repeats
@jorge @eliasp @tml @aeva @mairin Not often and itâs probably not something youâd often see off a VM cluster because real hardware doesnât vary in performance from run to run, but with hundreds of VMs being rebooted by devs and QA it happened throughout the day. Iâm sure the very occasional random hung boots I have seen on ubuntu are related.My new job uses docker for dev and systemd doesnât happen any more.
(DIR) Post #AX8ZbrqItUHBEPG21g by resuna@ohai.social
2023-06-28T03:18:45Z
0 likes, 0 repeats
@justinz @jorge @eliasp @tml @aeva @mairin It wasnât starting them all up at once. Developers and QA testers would start and stop them as they needed throughout the day. RHEL6 and Windows and whatever version of SUSE we were using had no problems. RHEL7 VMs could be relied upon to get the heebie-jeebies multiple times a day, and you could see the startup order changing in the logs.