Al Jazeera English is a pretty impressive news channel. Their journalists are pretty sharp and honest. There appears to us in the west to be some bias, but in reality I find them to be more impartial than any western news service. Head over to livestation.com and watch it. I think you'd be surprised at how good a job they do. This is not the same Al Jazeera that made their name parroting Iraqi propaganda during the recent American invasion there, for all intents and purposes.
With the satellites at 600 miles, and if they truly could cover the entire earth, they could provide internet access of some kind to the ISS. Would beat the current system of vnc over radio link.
Making it just as heavy as Gnome and KDE now?
on
Xfce 4.8 Released
·
· Score: 5, Interesting
Hopefully all these new-fangled frameworks and technologies aren't going to turn Xfce into just another Gnome or KDE competitor. Xfce was always fast and light. Hopefully it stays that way.
What's wrong with buy.com? I've bought thousands of dollars worth of things from them over the years and never had a problem with the wrong item or being charged differently than I expected. I've never had to return anything to them, so I can't speak to that.
It's also true that the air coming out of a tier-4 diesel engine is supposed to be cleaner than the air that went in, in terms of articulates and NOx. But that does very little to address the root causes of global climate change, which is net CO2.
If it does reduce dependency on foreign oil, it's a good thing, though, but it's certainly not about reducing CO2 emissions.
I always thought the point of CG was to make a scene that didn't look like CG. I think some movie directors would argue that if you notice the CG then the CG was done wrong. Some of the most spectacular scenes in LOTR had CG in them but it was a subtle blending of CG (massive amounts in battle scenes) with natural new zealand scenery, sets, models, and actors.
Brother Stallman has turned out to be right so many times that maybe he's a prophet by now. In particular I remember reading his chilling essay, "The Right to Read" and then watching as the Kindle and other DRM-ladened projects have made real what before was a comfortable, 1984-esque story. Very uncomfortable to see private companies (with purchased legislation) doing what Orwell thought governments would do.
Haha you are funny if you think that the free market will simply mean growers make more money. The reality is that growers are price takers, not price setters. And they don't get to take the price that the end consumer is willing to pay either. They get the price the large food companies are willing to pay. And because of farm subsidies, food companies are always guaranteed to find commodities elsewhere if the producer doesn't want to sell. The only leverage against this kind of market behavior is to have farms enter into cooperatives, as they have in times past.
The price of commodity goods does track the price of the final food products, but not very closely and always on the low side. High commodity prices immediately leads to high food prices, but when commodity prices drop, food prices do not fall very quickly. So what food scarcity really means is the middle food companies get really fat while the consumers and producers get shafted.
She certainly didn't drive her motorcycle through there, her father isn't a scientist, and she certainly isn't the only person to go there. She did, however, take the pictures. If I recall it turned out she just went on one of the many exclusion zone tours that have been going through there for years.
Sendmail has one redeeming feature: milters. Postfix is only now starting to support sendmail-compatible milter filters. The ability to filter and discard spam at the connection level is, my opinion, better and cleaner than hackish solutions like amavisd.
So what you are really saying is that a *single* external hard drive isn't a backup, which is true. But if a number of external drives are used in a rotation then I'd say that is a backup solution. And send an encrypted disk to a friend's house. And if you combine Time Machine with that you have both backup and an archive. I'd say it is cheaper and faster to do this than to rely on some cloud. That said, the cloud can certainly be a part of the solution, but I would not argue that the cloud is the only real backup solution.
In the case of this type of virus, a backup isn't what you need anyway. A corrupt file "backed" up to the backup is just as useless as it is on your computer's main drive. What folks need in this case in an archive. Backup is only good for restoring a failed drive. An archive lets you recover from file corruption, deletions, changes, etc. Two different but equally vital strategies.
GTK+ is certainly weaker on the Windows and Mac side than Qt. But as far as speed and ease of coding, GTK+ is right up there. GTK++ is a great binding for C++, and PyGTK is quite good too. Writing Qt GUIs in PyQT is okay, but it's essentially the same as writing C++ code. In fact, it is the bindings where Qt is the weakest and where GTK+ is a bit better. Qt is written in C++ using the C++ object model and all it's features and warts. This does not always translate very well to other languages. A few years ago all Qt bindings for other languages were based on the QtC bindings because of this. Now Qt bindings are better. But GTK+, being based on a much simpler object model (the GOject) is very easy to wrap in C++, Perl, Python, Ruby, etc. PyGTK is one of the more comfortable toolkits to develop in and feels more at home in Python than Qt (PyQT; never have used pyside).
GTK+ is hardly the "grave danger" the parent claims. For a lot of things it is a very nice toolkit to develop in. And the parent's statements on Tk are are not quite accurate. Tk is still alive and well, and looks reasonable in the modern incarnations on the big 3 platforms. It seems a bit daft to me to embed another entire language in my program (tcl) but sometimes that just might be the easiest. Many little utilities written in Python still use Tk. It's fast, easy, and always there if Python is there.
I disagree about not needing to feel the keys. Why is it called "touch typing" if touch is not involved?
As I type this now, I am positioning my fingers by muscle memory, yes, but the nice indent on the tops of the keys help my fingers find the center of the keys. Without them I might hit between keys. Also my forefingers sit on the keyboard and feel the little nubs that mark the home keys. This way I know I'm starting from the right position. All the while I'm not needing to look at the keyboard, but the feel definitely improves the accuracy and speed. I'd say that touch typists do need feel and some sort of feedback. That's why they call it "touch typing" after all. Both of these things are why touch typists in the past loved the IBM keyboard. It had a great and accurate feel, and excellent feedback (the click).
All this is is part of why I despise the chicklet keyboard. I lose all sorts of accuracy because it's much harder to quickly know if you're on the center of each key or not. The only benefit the chicklet keyboard has is the reduced amount of key travel. As far as reduced-travel keyboards go, Lenovo's laptop keyboards are by far the best.
I'm not quite a touch typist, but I can type pretty fast, and I have tried the iPad (lying flat on a desk of course), and I can type okay with it, but it's not great. Also it cannot tell the difference between resting my finger on a key and "pressing" a key. Makes typing very tiring. Back in the day they would teach typists to hold their wrists up, but even then the fingers could rest on the keys. Holding both hands off the surface except to "type" would seem to be exhausting.
I'm glad you are able to assume so much about my experience with XFS. I wouldn't have mentioned it all if I hadn't had at least some experience with it, recently (this year even). Give me at least that much credit please before patronizing me.
Anyway, as it happens I have run XFS on a 6 TB array until about two months ago on RHEL 5. I have no idea if RHEL5 has the latest kernel fixes that were mentioned in this thread for XFS. I lost data on two occasions to crashes in the last 2 years, but the bigger problem, and the problem that has turned me right off of XFS, is that *every* unintended reboot (crash) of my server resulted in me having to manually run fsck on the console and fix a corrupted file system. This was not even on a busy server. Maybe two users using files over Samba. Yet the file system needed manual repair. I have never ever had this happen with Ext3 (with light usage). Fortunately such crashes were rare. Anyway, somehow the fsck built into the init system couldn't ever repair it during the boot process. As near as I could tell this happened to me every time. The most recent crash was about a year ago. When it came time to replace the SAN, I just formatted the volume to Ext3 for now until I decide how to best proceed.
Perhaps as you say my understanding of ZFS is faulty. But I understand that ZFS is designed to make file truncation much more improbably because blocks are never re-written. Files are guaranteed to be, at any moment, intact and consistent. Only changes to the files (new blocks, copy-on-write) are lost. Is this not true? My experience seems to suggest that it is true in general. Sun certainly claims this. And in a paper from UW-M (zhang, Rajimwale, Arpaci-Dusseau, Arpaci-Dusseau) they say:
In our analysis, we find that ZFS is indeed robust to a wide range of disk corruptions, thus partially confirming that many of its design goals have been met. However, we also find that ZFS often fails to maintain data integrity in the face of memory corruption. In many cases, ZFS is either unable to detect the corruption, returns bad data to the user, or simply crashes. We further find that many of these cases could be avoided with simple techniques.
So ZFS is certainly prone to corruption, but more than likely runtime corruption (bad RAM, etc) than corruption of the FS due to power failure, even during busy times. Sun themselves claim to have pulled the power on busy SANs on numerous occasions and never had file system corruption, though as you say changes being made are lost, but I don't think open files were truncated.
Given my experience, if I can run the same ZFS version on Linux that I do on Solaris, I'll take it over XFS any day. At least for my particular usage needs. Given that I really don't like Solaris, and I don't care about ZFS specifically, I'm anxiously awaiting BtrFS on Linux. Maybe then I'll be a fan boy you can accuse of not knowing what I'm talking about.
Yes you can take your pick. Do you want fuel efficiency (less CO2) or emissions (more other things)? Pick one. And with diesels it's even worse. You can have great fuel efficiency and low particulate emission by running the engine hotter and at higher compression. But that generates NOx. Or you can recirculate exhaust and do weird cooling tricks and have zero NOx emissions, but worse fuel economy and more particulates because of lower combustion temperatures. Of course particulates at sea are not a real problem as they are just carbon substances and are pumped into the seawater directly usually.
This article is a classic example of folks going after the wrong things and, if they push legislation, having the exact opposite affect on the environment that they intended. Just like misguided EPA tier-4 diesel regulations. Will ultimately increase CO2 by quite a bit because of misguided legislation on emissions that are still nothing compared to cars.
In an enterprise you're typically dealing with SAN. Just simply "adding physical volumes" isn't quite so simple. What if your disk array is full? Just tack a USB disk on the server? For us, all our SANs are hardware RAID (we don't use RAID-Z), so adding new volumes, as you suggest, involves buying at least 4 disk (RAID-6), sticking them in the chassis and creating a hardware volume set. It's quite an undertaking to expand storage. LVM can certainly accommodate our hardware, but would certainly not be efficient use of our disks. LVM's need to have unallocated space for snapshots has always been a weak spot. LVM snapshots are actually writable, though (BtrFS and ZFS snapshots are read-only). ZFS snapshots may be slower, but they are easier and more flexible. Perhaps some of the speed issues with ZFS snapshots come from the idea of thousands of sub-file-systems. Some overhead there, but the flexibility is totally worth it.
Wrong answer. XFS is extremely prone to data corruption if the system goes down uncleanly for any reason. We may strive for nine nines, but stuff still happens. A power failure on a large XFS volume is almost guaranteed to lead to truncated files and general lost data. Not so on ZFS.
30 minutes? That's insane. An LVM2 snapshot would take seconds. I fail to see how that's not quick, and how "lvcreate -s" is less convenient.
Glad to know LVM is faster though. However, as I stated before it's not convenient. With ZFS I do the following things: - snapshot the works every night, and keep 7 days worth of snapshots. - some directories are snapshotted every night, but I keep 365 snapshots (one year). For example the directories that our financial folk use. - snapshot important directories every hour, keep 24 hours worth
You simply cannot do that with LVM. Sorry. How would I know how much free volume space to plan for? If I have a 10 TB disk, do I plan to use 6 TB of it and leave 4 TB for snapshots? Snapshots consume as much space as subsequent changes. For the 365 say snapshots, this could be a lot or very little depending on what has been touched.
I can't even make sense of these two sentences. What you're saying is, an LVM snapshot requires free space, and er, a ZFS snapshot requires free space?
It's very simple. LVM snapshots require free volume set space. If your volume group is 10 TB, then you must leave unallocated space on it for the snapshots to consume. On ZFS you don't need to do this. Any free space on the file system can be used for either files or snapshots; it's all the same pool. To do snapshots with LVM the way I do with ZFS would require me to set aside a lot of space. Very unefficient and wasteful.
As far as I can tell, BtrFS will work in a similar way to ZFS, bypassing the need for LVM. Which I'm totally okay with.
You mean like how the Nvidia GPU driver has failed because of licensing conflict? I see no reason why the ZFS module can't be distributed in a similar manner to the nvidia driver. I'm sure that rpmfusion could host binary RPMs without problem. They wouldn't be violating the GPL because it would be you the user who taints the kernel.
Of course ZFS on Linux probably isn't aimed at normal users anyway. It's far more likely to be used by people with existing ZFS infrastructure (large fiber-channel arrays, etc). In my opinion, ZFS on linux gives a smoother migration path away from Oracle Solaris and ZFS.
ZFS is, until BtrFS hits truly enterprise stable, the only FS for large disks, in my opinion. I currently run ZFS on about 10 TB. I never worry about a corrupt file system, never have to fsck it. And snapshots are cheap and fast. I shapshot the entire 10 TB array in about 30 minutes (about 2000 file systems). Then I back up from the snapshot. In other areas of the disk I do hourly snapshotting. Indeed snapshots are the kill feature for me for ZFS. LVM has snapshots, true, but they are not quick or convenient compared to ZFS. In LVM I can only snapshot to unused space in the volume set. With ZFS you can snapshot as long as you have free space. The integration of volume management and the file system may break a lot of people's ideas of clear separation between layers, but from the admin's point of view it is really nice.
We'll ditch ZFS and Solaris once BtrFS is ready. BtrFS is close, though; should work well for things like home servers, so try it out if you have a large MythTV system.
If you want to really save lives, crack down on seat belt use. Honestly, given the rate of increase of cell phone ownership and use, the statistics for highway fatalities have not gone up proportionally. Given that 10 years ago not that many people had cell phones and today everyone has one, you'd think listening to the distracted driving folks that the highways would be soaked in blood. That is demonstrably not the case.
Yes cell phones can cause accidents. Yes distracted driving can kill. But that doesn't mean we're having a huge epidemic. The only legal thing that might need to happen would be to help make it easier to win insurance judgements against accident instigators who were texting while driving.
A study by a not-so-unbiased industry group found in Canada that expressly banning texting or cell phone use while driving would simply lead to people hiding their phone use and cause more accidents. Stands to reason that if you're trying to use your phone and hide it from the police while driving, that you'd be a much worse driver.
Some regulation may well be needed; but completely going overboard with regulation or deregulation of anything seems to be a bad thing.
Been running Firefox 4 beta on Fedora 12 for months. Seems to just work on recent distros, provided you have GTK+ installed in the system. Just untarred the thing to a folder and ran the firefox binary inside.
We definitely need to rid ourselves of the "combustion is bad" mentality which is patently false and really stupid. When we finally get to the point where we can make a hydrocarbon molecule in a 100% carbon-neutral way (either from plants or some solar-powered catalytic process) then I don't see any problems with widespread use of IC engines in many applications. Electricity is nice and all, but simply doesn't cut it in many applications. Batteries just don't hold a candle to a gasoline or diesel molecule. Electric big rigs with batteries certainly aren't practical, and electrified rail can't run everywhere we need to haul stuff. If we had a renewable source of diesel fuel, with the new draconian diesel emissions regs coming into force, the air coming out of a big rig would be cleaner than the air going in, in terms of pollution. That wouldn't be a bad thing at all.
Several labs in my Uni's Chemistry Department regularly employ LASER choppers, if not "cutters". Simply stick the disk with the slits you want (to set the pulse duration) in front of the laser and set the motor to the desired RPM. That's how they get he pulse durations and frequencies that they need for their experiments. One could also use liquid crystal to turn the beam on and off rapidly. Come to think of it, I have no idea how a CD or DVD burner controls its laser. Maybe the laser can be turned on and off fast enough.
"Studies" apparently show lots of things. Most of the problems you mention are well-known, (even by us farmers), and are caused largely by unwise government subsidies. Obviously the farm bill is pushed by American farmers (of which I am not), and of course EU farmers, which is definitely short-sighted.
As for precision fertilizer, you are quite mistaken. Farmers are jumping at the bit to do this sort of thing, but so far it's just not economical yet. I can easily meter an average fert rate across my 52' drill, but doing individual runs is more complicated than you think. But the technology is coming soon, and I will definitely be adopting it when it's feasible, technologically and economically. I've seen a fertilizer bill as high as $250,000 for 2500 acres. Don't you think I'd like to reduce that? The primary motivation isn't shiny things, but reducing costs which almost always equals reducing environmental footprint.
Slashdot Mirror
Al Jazeera English is a pretty impressive news channel. Their journalists are pretty sharp and honest. There appears to us in the west to be some bias, but in reality I find them to be more impartial than any western news service. Head over to livestation.com and watch it. I think you'd be surprised at how good a job they do. This is not the same Al Jazeera that made their name parroting Iraqi propaganda during the recent American invasion there, for all intents and purposes.
With the satellites at 600 miles, and if they truly could cover the entire earth, they could provide internet access of some kind to the ISS. Would beat the current system of vnc over radio link.
Hopefully all these new-fangled frameworks and technologies aren't going to turn Xfce into just another Gnome or KDE competitor. Xfce was always fast and light. Hopefully it stays that way.
From what I heard, one of the Apollo 13 astronauts remarked that the movie was pretty accurate to the real event, save for the swearing.
What's wrong with buy.com? I've bought thousands of dollars worth of things from them over the years and never had a problem with the wrong item or being charged differently than I expected. I've never had to return anything to them, so I can't speak to that.
It's also true that the air coming out of a tier-4 diesel engine is supposed to be cleaner than the air that went in, in terms of articulates and NOx. But that does very little to address the root causes of global climate change, which is net CO2.
If it does reduce dependency on foreign oil, it's a good thing, though, but it's certainly not about reducing CO2 emissions.
I always thought the point of CG was to make a scene that didn't look like CG. I think some movie directors would argue that if you notice the CG then the CG was done wrong. Some of the most spectacular scenes in LOTR had CG in them but it was a subtle blending of CG (massive amounts in battle scenes) with natural new zealand scenery, sets, models, and actors.
Brother Stallman has turned out to be right so many times that maybe he's a prophet by now. In particular I remember reading his chilling essay, "The Right to Read" and then watching as the Kindle and other DRM-ladened projects have made real what before was a comfortable, 1984-esque story. Very uncomfortable to see private companies (with purchased legislation) doing what Orwell thought governments would do.
Haha you are funny if you think that the free market will simply mean growers make more money. The reality is that growers are price takers, not price setters. And they don't get to take the price that the end consumer is willing to pay either. They get the price the large food companies are willing to pay. And because of farm subsidies, food companies are always guaranteed to find commodities elsewhere if the producer doesn't want to sell. The only leverage against this kind of market behavior is to have farms enter into cooperatives, as they have in times past.
The price of commodity goods does track the price of the final food products, but not very closely and always on the low side. High commodity prices immediately leads to high food prices, but when commodity prices drop, food prices do not fall very quickly. So what food scarcity really means is the middle food companies get really fat while the consumers and producers get shafted.
She certainly didn't drive her motorcycle through there, her father isn't a scientist, and she certainly isn't the only person to go there. She did, however, take the pictures. If I recall it turned out she just went on one of the many exclusion zone tours that have been going through there for years.
Sendmail has one redeeming feature: milters. Postfix is only now starting to support sendmail-compatible milter filters. The ability to filter and discard spam at the connection level is, my opinion, better and cleaner than hackish solutions like amavisd.
So what you are really saying is that a *single* external hard drive isn't a backup, which is true. But if a number of external drives are used in a rotation then I'd say that is a backup solution. And send an encrypted disk to a friend's house. And if you combine Time Machine with that you have both backup and an archive. I'd say it is cheaper and faster to do this than to rely on some cloud. That said, the cloud can certainly be a part of the solution, but I would not argue that the cloud is the only real backup solution.
In the case of this type of virus, a backup isn't what you need anyway. A corrupt file "backed" up to the backup is just as useless as it is on your computer's main drive. What folks need in this case in an archive. Backup is only good for restoring a failed drive. An archive lets you recover from file corruption, deletions, changes, etc. Two different but equally vital strategies.
GTK+ is certainly weaker on the Windows and Mac side than Qt. But as far as speed and ease of coding, GTK+ is right up there. GTK++ is a great binding for C++, and PyGTK is quite good too. Writing Qt GUIs in PyQT is okay, but it's essentially the same as writing C++ code. In fact, it is the bindings where Qt is the weakest and where GTK+ is a bit better. Qt is written in C++ using the C++ object model and all it's features and warts. This does not always translate very well to other languages. A few years ago all Qt bindings for other languages were based on the QtC bindings because of this. Now Qt bindings are better. But GTK+, being based on a much simpler object model (the GOject) is very easy to wrap in C++, Perl, Python, Ruby, etc. PyGTK is one of the more comfortable toolkits to develop in and feels more at home in Python than Qt (PyQT; never have used pyside).
GTK+ is hardly the "grave danger" the parent claims. For a lot of things it is a very nice toolkit to develop in. And the parent's statements on Tk are are not quite accurate. Tk is still alive and well, and looks reasonable in the modern incarnations on the big 3 platforms. It seems a bit daft to me to embed another entire language in my program (tcl) but sometimes that just might be the easiest. Many little utilities written in Python still use Tk. It's fast, easy, and always there if Python is there.
I disagree about not needing to feel the keys. Why is it called "touch typing" if touch is not involved?
As I type this now, I am positioning my fingers by muscle memory, yes, but the nice indent on the tops of the keys help my fingers find the center of the keys. Without them I might hit between keys. Also my forefingers sit on the keyboard and feel the little nubs that mark the home keys. This way I know I'm starting from the right position. All the while I'm not needing to look at the keyboard, but the feel definitely improves the accuracy and speed. I'd say that touch typists do need feel and some sort of feedback. That's why they call it "touch typing" after all. Both of these things are why touch typists in the past loved the IBM keyboard. It had a great and accurate feel, and excellent feedback (the click).
All this is is part of why I despise the chicklet keyboard. I lose all sorts of accuracy because it's much harder to quickly know if you're on the center of each key or not. The only benefit the chicklet keyboard has is the reduced amount of key travel. As far as reduced-travel keyboards go, Lenovo's laptop keyboards are by far the best.
I'm not quite a touch typist, but I can type pretty fast, and I have tried the iPad (lying flat on a desk of course), and I can type okay with it, but it's not great. Also it cannot tell the difference between resting my finger on a key and "pressing" a key. Makes typing very tiring. Back in the day they would teach typists to hold their wrists up, but even then the fingers could rest on the keys. Holding both hands off the surface except to "type" would seem to be exhausting.
I'm glad you are able to assume so much about my experience with XFS. I wouldn't have mentioned it all if I hadn't had at least some experience with it, recently (this year even). Give me at least that much credit please before patronizing me.
Anyway, as it happens I have run XFS on a 6 TB array until about two months ago on RHEL 5. I have no idea if RHEL5 has the latest kernel fixes that were mentioned in this thread for XFS. I lost data on two occasions to crashes in the last 2 years, but the bigger problem, and the problem that has turned me right off of XFS, is that *every* unintended reboot (crash) of my server resulted in me having to manually run fsck on the console and fix a corrupted file system. This was not even on a busy server. Maybe two users using files over Samba. Yet the file system needed manual repair. I have never ever had this happen with Ext3 (with light usage). Fortunately such crashes were rare. Anyway, somehow the fsck built into the init system couldn't ever repair it during the boot process. As near as I could tell this happened to me every time. The most recent crash was about a year ago. When it came time to replace the SAN, I just formatted the volume to Ext3 for now until I decide how to best proceed.
Perhaps as you say my understanding of ZFS is faulty. But I understand that ZFS is designed to make file truncation much more improbably because blocks are never re-written. Files are guaranteed to be, at any moment, intact and consistent. Only changes to the files (new blocks, copy-on-write) are lost. Is this not true? My experience seems to suggest that it is true in general. Sun certainly claims this. And in a paper from UW-M (zhang, Rajimwale, Arpaci-Dusseau, Arpaci-Dusseau) they say:
So ZFS is certainly prone to corruption, but more than likely runtime corruption (bad RAM, etc) than corruption of the FS due to power failure, even during busy times. Sun themselves claim to have pulled the power on busy SANs on numerous occasions and never had file system corruption, though as you say changes being made are lost, but I don't think open files were truncated.
Given my experience, if I can run the same ZFS version on Linux that I do on Solaris, I'll take it over XFS any day. At least for my particular usage needs. Given that I really don't like Solaris, and I don't care about ZFS specifically, I'm anxiously awaiting BtrFS on Linux. Maybe then I'll be a fan boy you can accuse of not knowing what I'm talking about.
Yes you can take your pick. Do you want fuel efficiency (less CO2) or emissions (more other things)? Pick one. And with diesels it's even worse. You can have great fuel efficiency and low particulate emission by running the engine hotter and at higher compression. But that generates NOx. Or you can recirculate exhaust and do weird cooling tricks and have zero NOx emissions, but worse fuel economy and more particulates because of lower combustion temperatures. Of course particulates at sea are not a real problem as they are just carbon substances and are pumped into the seawater directly usually.
This article is a classic example of folks going after the wrong things and, if they push legislation, having the exact opposite affect on the environment that they intended. Just like misguided EPA tier-4 diesel regulations. Will ultimately increase CO2 by quite a bit because of misguided legislation on emissions that are still nothing compared to cars.
In an enterprise you're typically dealing with SAN. Just simply "adding physical volumes" isn't quite so simple. What if your disk array is full? Just tack a USB disk on the server? For us, all our SANs are hardware RAID (we don't use RAID-Z), so adding new volumes, as you suggest, involves buying at least 4 disk (RAID-6), sticking them in the chassis and creating a hardware volume set. It's quite an undertaking to expand storage. LVM can certainly accommodate our hardware, but would certainly not be efficient use of our disks. LVM's need to have unallocated space for snapshots has always been a weak spot. LVM snapshots are actually writable, though (BtrFS and ZFS snapshots are read-only). ZFS snapshots may be slower, but they are easier and more flexible. Perhaps some of the speed issues with ZFS snapshots come from the idea of thousands of sub-file-systems. Some overhead there, but the flexibility is totally worth it.
Wrong answer. XFS is extremely prone to data corruption if the system goes down uncleanly for any reason. We may strive for nine nines, but stuff still happens. A power failure on a large XFS volume is almost guaranteed to lead to truncated files and general lost data. Not so on ZFS.
Glad to know LVM is faster though. However, as I stated before it's not convenient. With ZFS I do the following things:
- snapshot the works every night, and keep 7 days worth of snapshots.
- some directories are snapshotted every night, but I keep 365 snapshots (one year). For example the directories that our financial folk use.
- snapshot important directories every hour, keep 24 hours worth
You simply cannot do that with LVM. Sorry. How would I know how much free volume space to plan for? If I have a 10 TB disk, do I plan to use 6 TB of it and leave 4 TB for snapshots? Snapshots consume as much space as subsequent changes. For the 365 say snapshots, this could be a lot or very little depending on what has been touched.
It's very simple. LVM snapshots require free volume set space. If your volume group is 10 TB, then you must leave unallocated space on it for the snapshots to consume. On ZFS you don't need to do this. Any free space on the file system can be used for either files or snapshots; it's all the same pool. To do snapshots with LVM the way I do with ZFS would require me to set aside a lot of space. Very unefficient and wasteful.
As far as I can tell, BtrFS will work in a similar way to ZFS, bypassing the need for LVM. Which I'm totally okay with.
You mean like how the Nvidia GPU driver has failed because of licensing conflict? I see no reason why the ZFS module can't be distributed in a similar manner to the nvidia driver. I'm sure that rpmfusion could host binary RPMs without problem. They wouldn't be violating the GPL because it would be you the user who taints the kernel.
Of course ZFS on Linux probably isn't aimed at normal users anyway. It's far more likely to be used by people with existing ZFS infrastructure (large fiber-channel arrays, etc). In my opinion, ZFS on linux gives a smoother migration path away from Oracle Solaris and ZFS.
ZFS is, until BtrFS hits truly enterprise stable, the only FS for large disks, in my opinion. I currently run ZFS on about 10 TB. I never worry about a corrupt file system, never have to fsck it. And snapshots are cheap and fast. I shapshot the entire 10 TB array in about 30 minutes (about 2000 file systems). Then I back up from the snapshot. In other areas of the disk I do hourly snapshotting. Indeed snapshots are the kill feature for me for ZFS. LVM has snapshots, true, but they are not quick or convenient compared to ZFS. In LVM I can only snapshot to unused space in the volume set. With ZFS you can snapshot as long as you have free space. The integration of volume management and the file system may break a lot of people's ideas of clear separation between layers, but from the admin's point of view it is really nice.
We'll ditch ZFS and Solaris once BtrFS is ready. BtrFS is close, though; should work well for things like home servers, so try it out if you have a large MythTV system.
If you want to really save lives, crack down on seat belt use. Honestly, given the rate of increase of cell phone ownership and use, the statistics for highway fatalities have not gone up proportionally. Given that 10 years ago not that many people had cell phones and today everyone has one, you'd think listening to the distracted driving folks that the highways would be soaked in blood. That is demonstrably not the case.
Yes cell phones can cause accidents. Yes distracted driving can kill. But that doesn't mean we're having a huge epidemic. The only legal thing that might need to happen would be to help make it easier to win insurance judgements against accident instigators who were texting while driving.
A study by a not-so-unbiased industry group found in Canada that expressly banning texting or cell phone use while driving would simply lead to people hiding their phone use and cause more accidents. Stands to reason that if you're trying to use your phone and hide it from the police while driving, that you'd be a much worse driver.
Some regulation may well be needed; but completely going overboard with regulation or deregulation of anything seems to be a bad thing.
Been running Firefox 4 beta on Fedora 12 for months. Seems to just work on recent distros, provided you have GTK+ installed in the system. Just untarred the thing to a folder and ran the firefox binary inside.
We definitely need to rid ourselves of the "combustion is bad" mentality which is patently false and really stupid. When we finally get to the point where we can make a hydrocarbon molecule in a 100% carbon-neutral way (either from plants or some solar-powered catalytic process) then I don't see any problems with widespread use of IC engines in many applications. Electricity is nice and all, but simply doesn't cut it in many applications. Batteries just don't hold a candle to a gasoline or diesel molecule. Electric big rigs with batteries certainly aren't practical, and electrified rail can't run everywhere we need to haul stuff. If we had a renewable source of diesel fuel, with the new draconian diesel emissions regs coming into force, the air coming out of a big rig would be cleaner than the air going in, in terms of pollution. That wouldn't be a bad thing at all.
Warning: Science content.
Several labs in my Uni's Chemistry Department regularly employ LASER choppers, if not "cutters". Simply stick the disk with the slits you want (to set the pulse duration) in front of the laser and set the motor to the desired RPM. That's how they get he pulse durations and frequencies that they need for their experiments. One could also use liquid crystal to turn the beam on and off rapidly. Come to think of it, I have no idea how a CD or DVD burner controls its laser. Maybe the laser can be turned on and off fast enough.
"Studies" apparently show lots of things. Most of the problems you mention are well-known, (even by us farmers), and are caused largely by unwise government subsidies. Obviously the farm bill is pushed by American farmers (of which I am not), and of course EU farmers, which is definitely short-sighted.
As for precision fertilizer, you are quite mistaken. Farmers are jumping at the bit to do this sort of thing, but so far it's just not economical yet. I can easily meter an average fert rate across my 52' drill, but doing individual runs is more complicated than you think. But the technology is coming soon, and I will definitely be adopting it when it's feasible, technologically and economically. I've seen a fertilizer bill as high as $250,000 for 2500 acres. Don't you think I'd like to reduce that? The primary motivation isn't shiny things, but reducing costs which almost always equals reducing environmental footprint.