You seem a bit confused. Those bugs are not something upstream can or should be looking in to. That would be like blaming the bash team for a bug in your script. Naturally each distro has their own different sysVinit package, each wants to boot a little differently.
As for the rest, Gentoo and Slackware are existing systemd-less distros. Currently, Jessie has it as optional, so creating a fork would be a matter or removing a few packages from the repo. If the GR doesn't pass, it might not stay that way, but that isn't a problem for the next 3 years.
Your point seems to be, "new stuff requires new debugging methods on failure, avoid new stuff."
My point is new things should HAVE an effective debugging technique.
I am fine with replacing or greatly updating SysV. I simply don't see systemd as an improvement on the balance. Its wrongs outweigh its rights.
As an aside, most competent admins wouldn't be "resorting" to the kludge you describe, and will probably just stare and you and blink rapidly when told doing so is supposed to be a good thing.
I guess they think the answer is to stay down while they do a total re-install without understanding what was wrong then?
The way you talk, it seems you are not only pleased to have systemd but deliriously happy that the people who don't want it may get stuck with it. Who pissed in your corn flakes?
As for the rest, TFA suggests that people are preparing to do the needed thing.
Even SysVinit isn't in such a hot state, it haven't made a release in five years, and the defacto upstream maintainers have been SUSE/Reed Hat for years. At some point they will drop maintaining it anymore.
So it hasn't actually needed a change in years (meaning it is fully matured) but you're worried that there isn't enough manpower to churn the code (that needs no churning)?!?
Why in the world would Xorg need systemd? It doesn't need it now.
SysV init (known as init for the rest of this post) is quite simple. It has just a few things that it is instructed to restart if it exit (the gettys for the system console and optionally serial console), and a display manager such as GDM. Then there's a line that makes it call rc.S with the runlevel whenever that changes. Beyond that, init just reaps any child processes that end up parented to it.
rc.S is a script that just looks at/etc/tc?.d where ? is the selected runlevel and runs the scripts that start with S in the order natural order (fixed by adding two digits so that S80foo starts before S81bar. It passes 'start' as the parameter to the script.
In addition, rc scripts starting with K are called with 'stop' as the parameter.
Done.
Should you want/need a parallel startup, modify rc.S (Debian Wheezy does that).
They already had a working init system. Not all new and shiny, but perfectly functional. They had the option to stick with it while advocating for a better choice to be created. Or pick just about anything that discourages the growing hairball.
In the past, I have resorted to booting with init=/bin/bash and then running the rc scripts by hand to see the problem. Systemd won't even try to work if it isn't pid1, and it can't be if I booted init=/bin/bash.
In other cases, I have booted to shell, mounted up the filesystems and then did/etc/init/d/ssh start so I could get a second opinion. Try that with systemd.
In any number of cases, I have had to set something up that the system scripts and configs didn't (and couldn't have) anticipated. It was a simple matter of editing a few init scripts...
Imagine the 'fun' if you need to boot to a rescue disk, chroot into the server filesystem and bring up services while you fix a problem.
These 'heroic' measures come up when a server is in a lights out environment hours away. Sometimes the best approach is to get it to limp along until regular hours.
Debian can already do parallel starts without systemd. That could be improved upon, but I see no reason systemd's approach is required for that.
Honest question, Can anyone out there name a single reason everything systemd does can't be done as well or better using a simple helper app to start the daemon (and optionally stick around to monitor/control it)?
Any clue why systemd should even have an interest in replacing the well tested nntpd?
When my mom needed something for email and web I gave her a machine I had with Linux on it. It has worked just fine even though she had very little experience using a computer.
And I can tell you that nerve damage (especially around the fingertips) is important to glove manufacturers, especially concerning sporting gloves, where the risk of such damage is high with or without gloves.
And yet, they manage to make functional sporting gloves without a 100,000% mark-up over the cost of a DIY version. The people working with the affordable prosthetics are also interested in not causing harm. They (and sporting glove manufacturers) just have a proper sense of proportion about it and a better understanding of the actual risks.
As for watchbands, I actually do know a few people who've had allergic reactions to watchbands of various kinds, starting with myself.
OMG, terminal wrist irritation. How long do you have? Oh, just changed watchbands and all's well? That sounds much smarter than getting a $10,000 FDA approved watchband.
That's what insurance is for.
That's a good one! You do know there's a lot of people with existing amputations who have no insurance, right?
By that time, the damage may already be permanent. That's one of the things that research would study before handing it off to an unsuspecting patient.
So, constriction bands around the viscera and the neck can be unregulated but a velcro strap around the wrist is likely to cause a permanent injury? B_U_L_L_S_H_I_T.
And I must say, one of us certainly doesn't get risk analysis but it isn't me. Risk analysis does not mean freaking out at even the barest hint of the possibility that someone may get a rash.
As for the FDA and risk analysis, they haven't a clue. They're the organization that doesn't get that short term terminal patients have already lost what they have to lose. At the same time, they approve drugs that can cause homicidal rage (which has resulted in deaths) as a barely effective aid to quitting cigarettes. I fully agree that for controlled poisons (drugs) and implantable devices some sort of regulation is required but frankly the FDA isn't it. The whole agency needs to be chopped up for parts and replaced.
It's fine to do that for gmail or yahoo, Comcast, etc but deepdarksecert.com might not appreciate it if iPhones are sending that information back to apple even if it is never published.
I have no idea about this device, but similar efforts in the same price range have been found to be more functional by their users than the expensive approved models.
That would require an even bigger violation. They would have to have the client send the actual configuration to Apple as well so they can have the data. Not all businesses would appreciate that.
I can tell you nobody has ever thought it was all that important with gloves and watchbands and we don't have a small army of people who were nerve damaged by their casio. I can tell you that if it costs $40,000 and you don't have that kind of cash laying around, it might as well not exist at all. Are you claiming people are better off with nothing? Are you willing to say that to their faces? Sorry, you're not rich enough to have a hand?
Or consider canes. If a cane is used improperly, it can cause back shoulder and arm pain. Should we make canes cost $40,000 or should we just adjust them differently if things start hurting?
Imagine the disaster it would be for the economy if we all had to wear only medically approved clothes complete with $40,000 belts and $100,000 shoes. But OMG, what if the belt fails and their pants fall and cause them to trip and trigger a nuclear meltdown, millions of lives are at stake here! $100,000 is such a small price to pay in order to safely not go naked in public!
I imagine the kid will do what the rest of us do. If the hand starts causing pain he'll use it less until it can be adjusted. Meanwhile, unlike before, he has a functional prosthetic hand.
I'll bet that the $500 beater is infinitely more useful than a Ferrari to someone who will never be able to afford a Ferrari.
In other words, that looks like about $39,955 worth of FUD (and unicorn hair). Most people really can't afford that much FUD. Thankfully, I'm not in the market for a prosthetic hand, but if I was, I would at least try the $45 one first.
Here we go. Please do tell us about how the "official" prosthetic costing $40,000 are totally not a ripoff even though they can be replaced by $45 printed prostetics because each one is hand carved by highly skilled gnomes from their own bones and tied together with unicorn hair and anything else will kill the wearer in the first 5 minutes.
Because then you are sending a lot of requests to random domains that may not be designed to handle the traffic? And a lot, a hell of a lot of mail servers out there for common email services use legacy mail servers not related to the domain of the email address (because the mail servers were set up before that particular email domain became popular).
So what magical thing is it that you think apple can do on it's servers that the mail client couldn't do for itself that somehow doesn't pester the domain the user enters?
Of course, if the MX for the domain falls over from 3 or 4 probes, you won't be getting any mail anyway.
No- not really. You see, as long as demand outstrips supply, the price goes up.
Kick the gearbox! This part wasn't about power, it's about consumer markets in general. Do you claim we just can't make enough TVs to satisfy demand? Car lots are empty because they can't get them in as fast as they sell? When is the last time you heard of anyone going to Best Buy and they were told "can't help you, our shelves are empty"? More than one consumer products manufacturer has been caught channel stuffing because production capability well outstripped demand.
As for bailouts, the whole world was screwed by the bubble bursting. Socialist or capitalist hardly mattered. However, by doubling down on the socialism (nationalizing the banks) and a fairly modest bailout loan, Iceland has recovered fairly quickly (more quickly than the U.S.). They also actually prosecuted fraudsters in their banking system. It is well understood that easing back on bank regulation in the early 2000's was a key factor that permitted the crisis.
Greece and Spain got sucked in by the austerity fad (started by a paper with a serious math error) and went that way rather than telling the banks to stuff it like Iceland did. Ireland suffered for a few years as they resisted taking the needed steps, but eventually liquidated the IBRC and got back on track.
Meanwhile, the U.S. proclaimed the crisis over by fiat, but it only ended for Wall Street. It continues to drag on (though with signs of improvement) for the rest of us.Our debt is rising. Iceland's is falling.
The part you're missing is that it costs nothing to throw it in the distro. That doesn't mean anyone is actually using it instead of going direct to ALSA most of the time.
The same distros also support legacy OSS, is that best of breed too? More likely, both are supported should the user run against a commercial app (like skype) that for some reason demands it.
I suppose it could be called best of breed among userspace sound daemons in the same sense that a saw with all but 1 tooth broken off is best of breed among saws with no teeth.
But distros hire actual professionals to make these decisions
Is that the same professionals who:
They probably should have waited another year before doing the changeover.
Apparently you don't mind 'smearing' all 'professionals', just the ones that are inconvieniant for you.
Slashdot Mirror
Was it really that hard to change the symlink name to enforce rpcbind coming after networking?
Don't confuse bugs in an individual distro's scripts for bugs in the underlying init maintained upstream.
You seem a bit confused. Those bugs are not something upstream can or should be looking in to. That would be like blaming the bash team for a bug in your script. Naturally each distro has their own different sysVinit package, each wants to boot a little differently.
As for the rest, Gentoo and Slackware are existing systemd-less distros. Currently, Jessie has it as optional, so creating a fork would be a matter or removing a few packages from the repo. If the GR doesn't pass, it might not stay that way, but that isn't a problem for the next 3 years.
Or that there are no known bugs to fix and haven't been in ages.
That's what happens when you follow the KISS principle.
Your point seems to be, "new stuff requires new debugging methods on failure, avoid new stuff."
My point is new things should HAVE an effective debugging technique.
I am fine with replacing or greatly updating SysV. I simply don't see systemd as an improvement on the balance. Its wrongs outweigh its rights.
As an aside, most competent admins wouldn't be "resorting" to the kludge you describe, and will probably just stare and you and blink rapidly when told doing so is supposed to be a good thing.
I guess they think the answer is to stay down while they do a total re-install without understanding what was wrong then?
The way you talk, it seems you are not only pleased to have systemd but deliriously happy that the people who don't want it may get stuck with it. Who pissed in your corn flakes?
As for the rest, TFA suggests that people are preparing to do the needed thing.
Even SysVinit isn't in such a hot state, it haven't made a release in five years, and the defacto upstream maintainers have been SUSE/Reed Hat for years. At some point they will drop maintaining it anymore.
So it hasn't actually needed a change in years (meaning it is fully matured) but you're worried that there isn't enough manpower to churn the code (that needs no churning)?!?
Why in the world would Xorg need systemd? It doesn't need it now.
SysV init (known as init for the rest of this post) is quite simple. It has just a few things that it is instructed to restart if it exit (the gettys for the system console and optionally serial console), and a display manager such as GDM. Then there's a line that makes it call rc.S with the runlevel whenever that changes. Beyond that, init just reaps any child processes that end up parented to it.
rc.S is a script that just looks at /etc/tc?.d where ? is the selected runlevel and runs the scripts that start with S in the order natural order (fixed by adding two digits so that S80foo starts before S81bar. It passes 'start' as the parameter to the script.
In addition, rc scripts starting with K are called with 'stop' as the parameter.
Done.
Should you want/need a parallel startup, modify rc.S (Debian Wheezy does that).
Except it's not. /usr/lib/systemd/systemd-journald /usr/lib/systemd/systemd-timesyncd /usr/lib/systemd/systemd-networkd
And those are all entirely independent of systemd?
They already had a working init system. Not all new and shiny, but perfectly functional. They had the option to stick with it while advocating for a better choice to be created. Or pick just about anything that discourages the growing hairball.
In the past, I have resorted to booting with init=/bin/bash and then running the rc scripts by hand to see the problem. Systemd won't even try to work if it isn't pid1, and it can't be if I booted init=/bin/bash.
In other cases, I have booted to shell, mounted up the filesystems and then did /etc/init/d/ssh start so I could get a second opinion. Try that with systemd.
In any number of cases, I have had to set something up that the system scripts and configs didn't (and couldn't have) anticipated. It was a simple matter of editing a few init scripts...
Imagine the 'fun' if you need to boot to a rescue disk, chroot into the server filesystem and bring up services while you fix a problem.
These 'heroic' measures come up when a server is in a lights out environment hours away. Sometimes the best approach is to get it to limp along until regular hours.
Debian can already do parallel starts without systemd. That could be improved upon, but I see no reason systemd's approach is required for that.
Honest question, Can anyone out there name a single reason everything systemd does can't be done as well or better using a simple helper app to start the daemon (and optionally stick around to monitor/control it)?
Any clue why systemd should even have an interest in replacing the well tested nntpd?
Even in the Debian TC that voted in the first place it ended in a 4 to 4 split with the chairman breaking the tie. That's not a minority, that's half.
Always threaten with a spoon.
When my mom needed something for email and web I gave her a machine I had with Linux on it. It has worked just fine even though she had very little experience using a computer.
And I can tell you that nerve damage (especially around the fingertips) is important to glove manufacturers, especially concerning sporting gloves, where the risk of such damage is high with or without gloves.
And yet, they manage to make functional sporting gloves without a 100,000% mark-up over the cost of a DIY version. The people working with the affordable prosthetics are also interested in not causing harm. They (and sporting glove manufacturers) just have a proper sense of proportion about it and a better understanding of the actual risks.
As for watchbands, I actually do know a few people who've had allergic reactions to watchbands of various kinds, starting with myself.
OMG, terminal wrist irritation. How long do you have? Oh, just changed watchbands and all's well? That sounds much smarter than getting a $10,000 FDA approved watchband.
That's what insurance is for.
That's a good one! You do know there's a lot of people with existing amputations who have no insurance, right?
By that time, the damage may already be permanent. That's one of the things that research would study before handing it off to an unsuspecting patient.
So, constriction bands around the viscera and the neck can be unregulated but a velcro strap around the wrist is likely to cause a permanent injury? B_U_L_L_S_H_I_T.
And I must say, one of us certainly doesn't get risk analysis but it isn't me. Risk analysis does not mean freaking out at even the barest hint of the possibility that someone may get a rash.
As for the FDA and risk analysis, they haven't a clue. They're the organization that doesn't get that short term terminal patients have already lost what they have to lose. At the same time, they approve drugs that can cause homicidal rage (which has resulted in deaths) as a barely effective aid to quitting cigarettes. I fully agree that for controlled poisons (drugs) and implantable devices some sort of regulation is required but frankly the FDA isn't it. The whole agency needs to be chopped up for parts and replaced.
It's fine to do that for gmail or yahoo, Comcast, etc but deepdarksecert.com might not appreciate it if iPhones are sending that information back to apple even if it is never published.
In this case, he must be really bad at milking since it was given free of charge.
I really have no idea what your quote has to do with anything, that tech is pure unobtanium, even at $40,000. The $40,000 model is nothing like that.
It's good that the work is being done, but it's not ready yet.
I have no idea about this device, but similar efforts in the same price range have been found to be more functional by their users than the expensive approved models.
That would require an even bigger violation. They would have to have the client send the actual configuration to Apple as well so they can have the data. Not all businesses would appreciate that.
I can tell you nobody has ever thought it was all that important with gloves and watchbands and we don't have a small army of people who were nerve damaged by their casio. I can tell you that if it costs $40,000 and you don't have that kind of cash laying around, it might as well not exist at all. Are you claiming people are better off with nothing? Are you willing to say that to their faces? Sorry, you're not rich enough to have a hand?
Or consider canes. If a cane is used improperly, it can cause back shoulder and arm pain. Should we make canes cost $40,000 or should we just adjust them differently if things start hurting?
Imagine the disaster it would be for the economy if we all had to wear only medically approved clothes complete with $40,000 belts and $100,000 shoes. But OMG, what if the belt fails and their pants fall and cause them to trip and trigger a nuclear meltdown, millions of lives are at stake here! $100,000 is such a small price to pay in order to safely not go naked in public!
I imagine the kid will do what the rest of us do. If the hand starts causing pain he'll use it less until it can be adjusted. Meanwhile, unlike before, he has a functional prosthetic hand.
I'll bet that the $500 beater is infinitely more useful than a Ferrari to someone who will never be able to afford a Ferrari.
In other words, that looks like about $39,955 worth of FUD (and unicorn hair). Most people really can't afford that much FUD. Thankfully, I'm not in the market for a prosthetic hand, but if I was, I would at least try the $45 one first.
Here we go. Please do tell us about how the "official" prosthetic costing $40,000 are totally not a ripoff even though they can be replaced by $45 printed prostetics because each one is hand carved by highly skilled gnomes from their own bones and tied together with unicorn hair and anything else will kill the wearer in the first 5 minutes.
Because then you are sending a lot of requests to random domains that may not be designed to handle the traffic? And a lot, a hell of a lot of mail servers out there for common email services use legacy mail servers not related to the domain of the email address (because the mail servers were set up before that particular email domain became popular).
So what magical thing is it that you think apple can do on it's servers that the mail client couldn't do for itself that somehow doesn't pester the domain the user enters?
Of course, if the MX for the domain falls over from 3 or 4 probes, you won't be getting any mail anyway.
No- not really. You see, as long as demand outstrips supply, the price goes up.
Kick the gearbox! This part wasn't about power, it's about consumer markets in general. Do you claim we just can't make enough TVs to satisfy demand? Car lots are empty because they can't get them in as fast as they sell? When is the last time you heard of anyone going to Best Buy and they were told "can't help you, our shelves are empty"? More than one consumer products manufacturer has been caught channel stuffing because production capability well outstripped demand.
As for bailouts, the whole world was screwed by the bubble bursting. Socialist or capitalist hardly mattered. However, by doubling down on the socialism (nationalizing the banks) and a fairly modest bailout loan, Iceland has recovered fairly quickly (more quickly than the U.S.). They also actually prosecuted fraudsters in their banking system. It is well understood that easing back on bank regulation in the early 2000's was a key factor that permitted the crisis.
Greece and Spain got sucked in by the austerity fad (started by a paper with a serious math error) and went that way rather than telling the banks to stuff it like Iceland did. Ireland suffered for a few years as they resisted taking the needed steps, but eventually liquidated the IBRC and got back on track.
Meanwhile, the U.S. proclaimed the crisis over by fiat, but it only ended for Wall Street. It continues to drag on (though with signs of improvement) for the rest of us.Our debt is rising. Iceland's is falling.
The part you're missing is that it costs nothing to throw it in the distro. That doesn't mean anyone is actually using it instead of going direct to ALSA most of the time.
The same distros also support legacy OSS, is that best of breed too? More likely, both are supported should the user run against a commercial app (like skype) that for some reason demands it.
I suppose it could be called best of breed among userspace sound daemons in the same sense that a saw with all but 1 tooth broken off is best of breed among saws with no teeth.
But distros hire actual professionals to make these decisions
Is that the same professionals who:
They probably should have waited another year before doing the changeover.
Apparently you don't mind 'smearing' all 'professionals', just the ones that are inconvieniant for you.