Quote:
Quote:
About timing, my slackware machine using systemd now boots in 2.3 seconds, compared to about 15 before. Whether or not that's significant enough for your tastes, why not do things correct and efficient, especially when it means Pat no longer has to maintain *any* init scripts, but instead just use upstream systemd files for the upstream authors? Quote:
Troll harder! "systemd is forcing functions from gnome3 and pulseaudio" is so wrong, it hurt my brain a little read. Cheers. |
Quote:
:hattip: |
Quote:
|
Heh! The Spock moment was excellent! :)
Sysvinit works fine for me. So did HAL and legacy GRUB, for that matter. I'm old. I don't like change. ;) |
Boot Time ????
I don't understand. What's with all this talk about boot times? I vaguely remember booting my main desktop system in July when I finished installing it. None of my encrypted machines (servers) at home have been rebooted this year at all. I have to use a cheat sheet to remind me of the start up security procedures because it's so long between reboots.
Do y'all reboot your production machines? Having battery backup, a rarely used small generator, and reliable power from the local electric company, I rarely lose electrical power (Northern California). No Tornadoes/Hurricanes, few Thunderstorms. So maybe my situation is not common. (However,in Alabama when the thunderstorms approached I shut the systems down for about an hour till the storm passed. Less downtime than than fixing broken systems.) One of the reasons I've used Slack since the 90's is stability. Once a new build is configured I leave it running for weeks at a time. It takes a lot longer to start up my workflow applications than it does to reboot, no matter how long the reboot takes. I use the machines at irregular times day and night and I don't want to spend 15 minutes setting everything up each time. My learning machines are always in hardware surgery having brain and memory transplants. They don't even successfully boot half the time after I get done with my Dr. Frankenstein-like kernel configurations. Their purpose is for me to learn, not necessarily for them to work. When they do manage to function I assign them tasks such as copying/testing disk drives for hours at a time. The boot times of these medical experiments isn't a factor. Just having them rise from the operating table occasionally is good enough. Boot times mattered when every change (even an IP address) in an MS Windows 98 production machine meant a reboot. Other than kernel changes, or unless one has a laptop and doesn't use hibernate, why would the boot time matter? I believe an earlier post referred to the boot time saved for them as the equivalent of a few sneezes a day. Good comparison! I'd like to broaden my view on this subject and am open to having my opinion on this matter change. Would someone please explain their situation where boot times matter. What's the faster boot time worth to you? What percentage of your annual income would you be willing to pay so that all of your computers would boot in 5 seconds instead of 60 seconds? (I thought I would get with the mood of the conversation and put a little opinion into my normally dry posts) |
Some people either care about the environment or care about their power bill and turn their computers off daily (or more than daily).
|
Quote:
|
Same here.
And as far as the "troll" comment on me goes... Son that ain't nothing I've heard before that gonna make me take a different color dump in my toilet than the day before.:rolleyes: |
I hope that peoples complaining about how evil systemd is, have tested systemd on a extended period. I hope too that they read the blog post by Lennart that explains A LOT, what is the goal of systemd, "how to do this", what are the different new concept that use systemd, what is the list of new cool features, etc...
Right now, it just seems to be a lot of hate about it just because it's Lennart, because it does not use shell script, it has dbus has a hard dependency and it try to replace some other component of a normal GNU/Linux distribution. Read all the documentation about systemd to have a better understanding of what it is, and then if you still hate it, you have the right to do so. If i'm mistaken and everyone ranting about systemd actually tested it toughtfully and read the real doc by Lennart (no bashing about his personality, he may be an ass, it does not change the fact that he can still have nice ideas.), I apologize sincerely for this complete out of subject post. |
Quote:
By this, I mean, I am arguing the case against systemd not based on its quality of code or functionality, or features, but because it breaks existing solutions and reduces the UNIX-like portability at the outset, and it is relatively unproven software trying to replace a proven, stable one. I am not bothered about the "merits" until the preliminary objections are addressed. Still going by the courtroom analogy, as I think the preliminary objections haven't been replied to satisfactorily, there is no need to get into the "merits" of the case. I ask your Honour to dismiss the suit with exemplary costs. :p |
Power Usage
Quote:
I don't run my air conditioning unless the temps approach 90 F inside my house (dry California, not humid). At that temperature I worry about my computers and I start sweating on the paperwork at my desk. I feel justified burning the electricity for the cooling at that point. I guess that's my compromise. Save electricity on air conditioning but keep the computers running. It's amazing the computers keep working although I run them at high temperatures in the summer time. I built all the boxes with great air flow. My project for next spring is to reduce the amount of server hardware in use by my paying clients and install a separate air conditioner just for a separate computer room/closet and my main home office. The extra hardware now consumes more power and generates more heat and noise. You didn't mention boot times in your response. Perhaps you had commented in an earlier thread about this. |
Quote:
|
Quote:
Think about Linux as living thing, it's always changing and evolving. People write good software, more people pick it up and install it, it becomes part of a distro and it stays until something better comes along and replaces it. If someone writes bad software, people vote with their feet (Unity, harhar). Lots of software is written to survive in this mode of living. It's modular, so it can live by itself or with few dependencies on things outside of a typical environment. It doesn't need weird and hard to understand software to coexist with to survive at all. Lots of software which embodies these ideas are things that people like, things which aren't bloated and unwieldy (GNOME begs to be a omni-punching-bag here). If a typical Linux software flow is of the evolutionary strategy, then stuff of the systemd sort feel a lot like something akin to creationism. Before you get your guns out for *that* battle, this isn't the place. Point being, systemd didn't battle it out among the other inits over years and come out on top, it was passed down to us from the RedHat cloud by an author with an attitude. Therefore, it's perfectly reasonable to be skeptical of something of that sort. If anything, it's surprising that more people aren't, but that's their own fault. Moving on... The sanest thing to do, if everyone else is insane, is to stay sane. No'm'say'n'? If every other distro is going to go wild with systemd, then it's extra important for Slackware and Slackware users to learn about it and pay attention to how it survives (or not) in the wild until it's good and ready to be decided on with level and slack brains. :) |
Here is a developer's opinion on using systemd
https://bbs.archlinux.org/viewtopic....49530#p1149530 |
Well, not only is Gnu Not Unix but Linux Is Not UniX either. (Hard to make the final X come out to a useful word.)
|
All times are GMT -5. The time now is 02:19 PM. |