You seem to be suggesting that snapshotting should be something that can be done on an arbitrary block device. To do this, you would have to have this supported in the block device handling
Well duh. What do you think an LVM virtual device is? Right, an instance of a block driver. Now what do you think a "physical" block device is? Right again, an instance of a block driver. Spot the difference? No, neither can I. Get my point yet? Ah, darn.
Do you have a copy of StarOffice from the mid-to-late 90's? Try running that in Linux now.
Huh? What makes you think it will not work? I must admit I do not have one of those sitting around, but if I did I would fully expect it to work. It would startle me if it did not, and I would probably do something about it if it did not. Oh, and could you please try keeping your future posts to verifiable facts?
just in terms of video based on where the video drivers run, performance will always be a contention when compared to OSes that structured so that the main video drivers are kernel or user/kernel mode hybrids as in Vista
Nonsense. Modern video hardware is predominantly driven by DMA, which requires an insignifcant number of kernel calls after initial setup. The rest of your points are just as empty and/or misinformed as your first, not worth a response.
And that is exactly my point: there is no need for this restriction. It is a stupid restriction that results from a stupid design. OK, at the time the design didn't seem so stupid because it could do things we could not do without it. But now, nearly ten years later, fossilized and frozen in time, it is obviously stupid and limiting.
You can't snapshot anything that isn't a LVM logical volume
True, and what kind of sense does that make? It is purely an artifact of the incumbent low level LVM model. Please go back and read the original post and notice how you misinterpreted it.
A snapshot works by creating a copy of the device, with the contents it had when the snapshot was created. If you make a snapshot of/dev/hda at 12:15, then you'll get/dev/mapper/snapshotted-hda as it was at 12:15, while/dev/hda will continue being possible to modify... Why would you change anything over?
Because with the incumbent volume management strategy you may not continue to use/dev/hda directly when it is snapshotted. You must access/dev/hda through some other device and that some other device must located in the/dev/mapper directory. No wonder you apparently mixed up what is a snapshot and what is being snapshotted - the way we currently do this in Linux is quite unnatural and is a wide open invitation to such confusion, not to mention a pointless makework project for system administrators.
The best solution would be for the Linux Kernel project to say, "Open source developers can do as they please, but we here at the Kernel project encourage developers to contribute to THESE specific projects: Gnome, Open Office, etc...
That is not going to happen, but if it did it would not include Gnome.
Lovely biting sarcasm aside, to be honest, our storage layering in Linux leaves much to be desired. As witness the slow pace of improvement of the volume manager in recent years. This does not prove that layering is bad, but it suggests that our current conception of layering sucks pretty badly. For example, we are burdened with a ridiculously complex interface between application programs and kernel-level volume management support. Managed volumes live off in their own name space. Why can't I say "/dev/hda, you are now snapshotted, shazam"?. No, instead I have change my system over to use/dev/mapper/snapshotted-hda or some such nonsense. Similarly, we are unable to manage all block devices using the same administration interface. No higher level raid integrated with the volume manager, instead this is a separate subsystem that fights a lot with LVM. Partition support hopelessly misfactored and broken. It goes on and on. Nothing unfixable but lots of unfixed brokenness. Compared to this mess, Sun's massive layering violations seem like a breath of fresh air.
But the thing to do is fix our broken implementation of layering and not be fooled into thinking that layers are bad. What is bad is exactly as the author here claims: it is bad to have no powerful capability to cross layer boundaries so that applications see a simple, powerful model instead of the current situation, where one's face is constantly rubbed in the minutae of layering administrivia. ZFS actually has layering, it just bypasses some traditional Unix subsystems and takes care of the functionality itself. But is wrong to conclude that this must therefore be the optimal approach just because it improves on the mess that preceded it. If ZFS internal interfaces are worth using, then they are worth using as core interfaces, not ZFS-only interfaces. Translated into Linux terms, the implication is that it is high time to get busy and rectify some of the serious deficiencies in our storage model. Not by mashing all the layers together, but by teaching them how to get along more efficiently and powerfully, and not be so layered that important things don't even work.
Note: perhaps the biggest design distinction between Linux and other Unixen is that, internally Linux is all just one big flat function space where anything can call anything else and share any data. This is said to be a reason why Linux is more efficient than, say, the Mach kernel with its microkernel layering. If being all one big hairball of functions is good for memory management, vfs, scheduling and so on, then why is it not also good for volume management? I don't know the answer to this, but I do know that we have plenty of bogus layering in our storage stack that has really slowed progress in recent years and needs a good dunging out. Any nonbogus layering can stay.
most home networks are not even 100Mbit, for the average joe user, who has multiple computers, it is much easier to set up an unsecured wireless network (802.11b/g), neither of these is much faster than 6-12mbits (in any situation I have seen)
Good point, however 1) 5-10 Mbit/sec wireless is still typically an order of magnitude faster than the "broadband" link and 2) nearly everybody has four or so wired ports on their wireless router. Once home users get into video and high res photos they quickly discover that wireless is just too painful compared to plugging in a physical wire. Now, home users with enough clue and energy to wire their homes properly will be a small minority for some time to come. But even a small minority of computer-enabled homes these days translates into millions. So I predict that the market for easy-to-use home servers will quickly climb into the millions of units. This has already started.
A server in every home? People want less to deal with, not more. Why have a home server if you can just connect to one online?
One word: privacy. Oh, another word: performance. Most home networks are 100 Mbits/sec while the internet connection is typically less than 1 Mbit/sec, two orders of magnitude less. This relative difference will remain for the forseeable future as home networks move to GigE while broadband speeds slowly increase into the tens of megabits. Think high resolution photos and video files.
Considering that GNOME is the default on suse, it is amazing. It looks like the more that the distros push GNOME, they more that they shoot themselves in the foot. Hopefully, this survey will stop that crap, but I am guessing that Novell will disregard this part.
Obviously, Novell pushes Gnome because Novell does not have nearly the same level of control of KDE as they do of Gnome, where a number of Gnome poster boys are on payroll. Ahem, and it is no exaggeration that Microsoft is now paying part of those Gnome paychecks. Odd.
To be honest I find it more than a little amusing that Microsoft has chosen to fund (with its "patent" deal) development work on a Linux desktop system. However much I prefer the other desktop, I do not deny that competition between desktops has value. What I do not like is Novell tilting the playing field, turning it more into a political competition and deep pockets competition than an honest features, performance and usability competition.
I am rather cross with you for lending your verbal support to Novell/Ximian's ignoble agenda to push KDE out of its well deserved and much loved position as the default desktop of the SuSE distribution. Which as you will recall, sparked a veritable customer revolt. How arrogant of you and those others involved. No, contrary to your claim [rlove.org], Ximian's offering is the second best Linux desktop, in my opinion. To be honest, I find Ximian's desktop downright irritating, clumsy, limited, buggy and really badly coded (yes I have been in there). However it is a fact that you have supported and continue to be an apologist for that cynical attempt to push KDE out of SuSE, that is not in doubt.
What is offtopic about that? It is a reasoned and honest response to the claim made on this page, linked from this article. On topic I say. And somebody with mod points yielded to the temptation to play the role of censor.
the sudden cratering of ResierFS demonstrates the huge Bus Factor of Open source software. Takeout one guy and major part of an operating system can suddenly become unsupported
ReiserFS (4) has not cratered, far from it. Instead, volunteers have stepped up to continue preparing the code for mainline merging. And far from showing a weakness, this sad tale proves convincingly that the code will carry on regardless of the misadventures of its original creators.
By the way, I for one favor merging Reiser 4 earlier rather than later. It includes a number of significant advances in filesystem design that we need in Linux, for example the dancing trees algorithm most probably is more efficient than Ext3's journalling, and even the chance of that means we need to get it in, shake it out, and prove it one way or the other. As for stability, the way to get stable is to achieve "good enough for the brave" then get lots of brave people to pitch in to help push it the rest of the way. Being in mainline makes it considerably easier for those brave people to configure it on and take for a test drive, not to mention help out. That is the way Linux has always worked, and most probably always will.
I am rather cross with you for lending your verbal support to Novell/Ximian's ignoble agenda to push KDE out of its well deserved and much loved position as the default desktop of the SuSE distribution. Which as you will recall, sparked a veritable customer revolt. How arrogant of you and those others involved. No, contrary to your claim, Ximian's offering is the second best Linux desktop, in my opinion. To be honest, I find Ximian's desktop downright irritating, clumsy, limited, buggy and really badly coded (yes I have been in there). However it is a fact that you have supported and continue to be an apologist for that cynical attempt to push KDE out of SuSE, that is not in doubt.
"AJAX stopped XAML dead in its tracks, at a time when new Microsoft technologies were still considered unstoppable. Why should we believe that Silverlight will be any different?"
as someone who likes money and lifestyle, who likes eating, who likes playing, who I choose to go where the money is. Idealism is fine. Personal crusades of windmill tilting are fine, Don. But, in the end, they don't put food on the table, they don't pay for the Hummer, they don't allow you to jet to Fiji for a cool vacation
Really? Speak for yourself. Some of us are able to remain true to our morals and live well at the same time.
Just tried the latest kernel and it hangs on trying to fire up the second ATA instance. Not even a kernel oops, nothing. That's true whether I use the vanilla kernel or Red Hat's RPM. Something is screwed up, and from the sounds of it, there's more than one of us experiencing a failure at the same point, so that would be the obvious suspect.
Which funds initiatives that try to "cure" AIDS by paying for expensive, patented drug treatments for some people, instead of cheap generic equivalents that could reach a much greater portion of those who need the treatments.
While this will help many people with AIDS, it will also support the concept of Intellectual Property, which is central to those billions of dollars that Bill Gates has invested in Microsoft.
And those not helped? Essentially a death sentence. For that reason and others I for one remained unconvinced that the Gates' foundation primary purpose is truly charity, as opposed to a glorified tax shelter and a means of extending influence.
Slashdot Mirror
You seem to be suggesting that snapshotting should be something that can be done on an arbitrary block device. To do this, you would have to have this supported in the block device handling
Well duh. What do you think an LVM virtual device is? Right, an instance of a block driver. Now what do you think a "physical" block device is? Right again, an instance of a block driver. Spot the difference? No, neither can I. Get my point yet? Ah, darn.
The original troll (err sorry, poster) appears to be completely unaware of what Unix versioned libraries are all about.
Do you have a copy of StarOffice from the mid-to-late 90's? Try running that in Linux now.
Huh? What makes you think it will not work? I must admit I do not have one of those sitting around, but if I did I would fully expect it to work. It would startle me if it did not, and I would probably do something about it if it did not. Oh, and could you please try keeping your future posts to verifiable facts?
just in terms of video based on where the video drivers run, performance will always be a contention when compared to OSes that structured so that the main video drivers are kernel or user/kernel mode hybrids as in Vista
Nonsense. Modern video hardware is predominantly driven by DMA, which requires an insignifcant number of kernel calls after initial setup. The rest of your points are just as empty and/or misinformed as your first, not worth a response.
Ok, under LVM, /dev/hda is a physical volume...
And that is exactly my point: there is no need for this restriction. It is a stupid restriction that results from a stupid design. OK, at the time the design didn't seem so stupid because it could do things we could not do without it. But now, nearly ten years later, fossilized and frozen in time, it is obviously stupid and limiting.
You can't snapshot anything that isn't a LVM logical volume
True, and what kind of sense does that make? It is purely an artifact of the incumbent low level LVM model. Please go back and read the original post and notice how you misinterpreted it.
You don't seem to understand snapshots
:-)
/dev/hda at 12:15, then you'll get /dev/mapper/snapshotted-hda as it was at 12:15, while /dev/hda will continue being possible to modify... Why would you change anything over?
/dev/hda directly when it is snapshotted. You must access /dev/hda through some other device and that some other device must located in the /dev/mapper directory. No wonder you apparently mixed up what is a snapshot and what is being snapshotted - the way we currently do this in Linux is quite unnatural and is a wide open invitation to such confusion, not to mention a pointless makework project for system administrators.
If you say so
A snapshot works by creating a copy of the device, with the contents it had when the snapshot was created. If you make a snapshot of
Because with the incumbent volume management strategy you may not continue to use
The best solution would be for the Linux Kernel project to say, "Open source developers can do as they please, but we here at the Kernel project encourage developers to contribute to THESE specific projects: Gnome, Open Office, etc...
That is not going to happen, but if it did it would not include Gnome.
Lovely biting sarcasm aside, to be honest, our storage layering in Linux leaves much to be desired. As witness the slow pace of improvement of the volume manager in recent years. This does not prove that layering is bad, but it suggests that our current conception of layering sucks pretty badly. For example, we are burdened with a ridiculously complex interface between application programs and kernel-level volume management support. Managed volumes live off in their own name space. Why can't I say "/dev/hda, you are now snapshotted, shazam"?. No, instead I have change my system over to use /dev/mapper/snapshotted-hda or some such nonsense. Similarly, we are unable to manage all block devices using the same administration interface. No higher level raid integrated with the volume manager, instead this is a separate subsystem that fights a lot with LVM. Partition support hopelessly misfactored and broken. It goes on and on. Nothing unfixable but lots of unfixed brokenness. Compared to this mess, Sun's massive layering violations seem like a breath of fresh air.
But the thing to do is fix our broken implementation of layering and not be fooled into thinking that layers are bad. What is bad is exactly as the author here claims: it is bad to have no powerful capability to cross layer boundaries so that applications see a simple, powerful model instead of the current situation, where one's face is constantly rubbed in the minutae of layering administrivia. ZFS actually has layering, it just bypasses some traditional Unix subsystems and takes care of the functionality itself. But is wrong to conclude that this must therefore be the optimal approach just because it improves on the mess that preceded it. If ZFS internal interfaces are worth using, then they are worth using as core interfaces, not ZFS-only interfaces. Translated into Linux terms, the implication is that it is high time to get busy and rectify some of the serious deficiencies in our storage model. Not by mashing all the layers together, but by teaching them how to get along more efficiently and powerfully, and not be so layered that important things don't even work.
Note: perhaps the biggest design distinction between Linux and other Unixen is that, internally Linux is all just one big flat function space where anything can call anything else and share any data. This is said to be a reason why Linux is more efficient than, say, the Mach kernel with its microkernel layering. If being all one big hairball of functions is good for memory management, vfs, scheduling and so on, then why is it not also good for volume management? I don't know the answer to this, but I do know that we have plenty of bogus layering in our storage stack that has really slowed progress in recent years and needs a good dunging out. Any nonbogus layering can stay.
most home networks are not even 100Mbit, for the average joe user, who has multiple computers, it is much easier to set up an unsecured wireless network (802.11b/g), neither of these is much faster than 6-12mbits (in any situation I have seen)
Good point, however 1) 5-10 Mbit/sec wireless is still typically an order of magnitude faster than the "broadband" link and 2) nearly everybody has four or so wired ports on their wireless router. Once home users get into video and high res photos they quickly discover that wireless is just too painful compared to plugging in a physical wire. Now, home users with enough clue and energy to wire their homes properly will be a small minority for some time to come. But even a small minority of computer-enabled homes these days translates into millions. So I predict that the market for easy-to-use home servers will quickly climb into the millions of units. This has already started.
A server in every home? People want less to deal with, not more. Why have a home server if you can just connect to one online?
One word: privacy. Oh, another word: performance. Most home networks are 100 Mbits/sec while the internet connection is typically less than 1 Mbit/sec, two orders of magnitude less. This relative difference will remain for the forseeable future as home networks move to GigE while broadband speeds slowly increase into the tens of megabits. Think high resolution photos and video files.
Considering that GNOME is the default on suse, it is amazing. It looks like the more that the distros push GNOME, they more that they shoot themselves in the foot. Hopefully, this survey will stop that crap, but I am guessing that Novell will disregard this part.
Obviously, Novell pushes Gnome because Novell does not have nearly the same level of control of KDE as they do of Gnome, where a number of Gnome poster boys are on payroll. Ahem, and it is no exaggeration that Microsoft is now paying part of those Gnome paychecks. Odd.
To be honest I find it more than a little amusing that Microsoft has chosen to fund (with its "patent" deal) development work on a Linux desktop system. However much I prefer the other desktop, I do not deny that competition between desktops has value. What I do not like is Novell tilting the playing field, turning it more into a political competition and deep pockets competition than an honest features, performance and usability competition.
I am rather cross with you for lending your verbal support to Novell/Ximian's ignoble agenda to push KDE out of its well deserved and much loved position as the default desktop of the SuSE distribution. Which as you will recall, sparked a veritable customer revolt. How arrogant of you and those others involved. No, contrary to your claim [rlove.org], Ximian's offering is the second best Linux desktop, in my opinion. To be honest, I find Ximian's desktop downright irritating, clumsy, limited, buggy and really badly coded (yes I have been in there). However it is a fact that you have supported and continue to be an apologist for that cynical attempt to push KDE out of SuSE, that is not in doubt.
What is offtopic about that? It is a reasoned and honest response to the claim made on this page, linked from this article. On topic I say. And somebody with mod points yielded to the temptation to play the role of censor.
the sudden cratering of ResierFS demonstrates the huge Bus Factor of Open source software. Takeout one guy and major part of an operating system can suddenly become unsupported
ReiserFS (4) has not cratered, far from it. Instead, volunteers have stepped up to continue preparing the code for mainline merging. And far from showing a weakness, this sad tale proves convincingly that the code will carry on regardless of the misadventures of its original creators.
By the way, I for one favor merging Reiser 4 earlier rather than later. It includes a number of significant advances in filesystem design that we need in Linux, for example the dancing trees algorithm most probably is more efficient than Ext3's journalling, and even the chance of that means we need to get it in, shake it out, and prove it one way or the other. As for stability, the way to get stable is to achieve "good enough for the brave" then get lots of brave people to pitch in to help push it the rest of the way. Being in mainline makes it considerably easier for those brave people to configure it on and take for a test drive, not to mention help out. That is the way Linux has always worked, and most probably always will.
Hi Robert,
I am rather cross with you for lending your verbal support to Novell/Ximian's ignoble agenda to push KDE out of its well deserved and much loved position as the default desktop of the SuSE distribution. Which as you will recall, sparked a veritable customer revolt. How arrogant of you and those others involved. No, contrary to your claim, Ximian's offering is the second best Linux desktop, in my opinion. To be honest, I find Ximian's desktop downright irritating, clumsy, limited, buggy and really badly coded (yes I have been in there). However it is a fact that you have supported and continue to be an apologist for that cynical attempt to push KDE out of SuSE, that is not in doubt.
Regards,
Daniel
WikiPedia uses Mono for its search facilities. The indexing and the actual searching is done by Mono-based applications.
Wow, Wikipedia uses mono for search? Interesting, if true. Search is the one aspect of Wikpedia that truly sucks. What should I learn from that?
"AJAX stopped XAML dead in its tracks, at a time when new Microsoft technologies were still considered unstoppable. Why should we believe that Silverlight will be any different?"
Um, Microsoft is the inventor of AJAX.
How deliciously ironic!
Hi Miguel, long time no see.
I suggest you call it nucleosis. You know what I mean.
Have a nice day.
as someone who likes money and lifestyle, who likes eating, who likes playing, who I choose to go where the money is. Idealism is fine. Personal crusades of windmill tilting are fine, Don. But, in the end, they don't put food on the table, they don't pay for the Hummer, they don't allow you to jet to Fiji for a cool vacation
Really? Speak for yourself. Some of us are able to remain true to our morals and live well at the same time.
Just tried the latest kernel and it hangs on trying to fire up the second ATA instance. Not even a kernel oops, nothing. That's true whether I use the vanilla kernel or Red Hat's RPM. Something is screwed up, and from the sounds of it, there's more than one of us experiencing a failure at the same point, so that would be the obvious suspect.
This problem needs to go to lkml, and cc Andrew.
This is easily the best article I've seen comparing data compression software
Hardly exhaustive. There is no mention of rzip.
I used to back in the day, but DHCP isn't as flaky anymore
But it is still flaky.
Which funds initiatives that try to "cure" AIDS by paying for expensive, patented drug treatments for some people, instead of cheap generic equivalents that could reach a much greater portion of those who need the treatments.
While this will help many people with AIDS, it will also support the concept of Intellectual Property, which is central to those billions of dollars that Bill Gates has invested in Microsoft.
And those not helped? Essentially a death sentence. For that reason and others I for one remained unconvinced that the Gates' foundation primary purpose is truly charity, as opposed to a glorified tax shelter and a means of extending influence.
they might get pissed if they get this "open" Office thingy on their computer that saves documents that they can't open up at school or work.
.doc format, and the real thing (Openoffice) is just a click away.
Rubbish, Openoffice saves perfectly well in Microsoft's
It would be extremely foolish to make business decisions with known bad data.
It would be extremely foolish for anybody to listen to a thing you say.
The problem of having a disproportionate polling set is you can estimate a revenue of the people you polled, at the risk of alienating millions more.
Oh come now. Nobody is going to be alienated by being offered the choice of free software on their new computer.