Doesn't this sort of application really not have much of a real world use? The trying out of a new kernel is okay, but you should be doing that on non-production machines anyhow. And as for cracker protection, chroot works better than usermode linux IMHO. I can see the development testing as being somewhat benficial, but that doesn't really test out the whole kernel does it? The only way to be sure that a kernel truly works is to boot it.
These opinions are my own, and not those of my alternate personalities. As such, any litigation should be directed at me sdalj; NOT ME; lkasdlf The dogs told me to do it. Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Software and Operating system evaluation is a very complex subject, and by no means is it ever black and white. There are many different ideologies that pervade the field. The international security community is embracing a new system know as Common Criteria(CC). The concept is that you define a set of objectives, and a CC testing facility checks to make sure that your software/OS/Hardware meets those standards. This is much more flexible than the TCSEC(Orange Book) evaluations. This also adds layers of complexity. Which CC eval spec do you need/want, who defines these specs, and how do you get your software tested. Well, the International Information Systems Security Certification Consortium (ISC)^2 for short has many resources for you to find CC specs and CC testing facilities. They also provide a comprehensive training and certification program for people interested in learning about information security. Web Site. The cert is well accepted, but don't think it's going to be a Microsoft MCP exam easy lick - 6 Hours, 250 questions.
Hope this is helpful. Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
The US was founded on the idea that people from all over the world and become patriated if they offered value to the country. If we deport all of these highly skilled workers, not only will we be hamstringing our Internet economy, but we will be going against some of the basic priciples of our country. Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Truthfully, there has to be innovation in the way that cases are being made anyhow. I know that there are many of us in the IT field who run into problems installing X equipment in Y case due to some problem or another. The major concern has to be that vendors follow a set STANDARD (psst... Microsoft, learn this term) that allows all equipment to fit all motherboards/cases/power supplies. I keep running in to cases where I have remove the internal drive bay assembly to fit RAM in with X mother board, or I have to purchase X adapter tray to make Y drive fit Z case. PCs are supposed to be modular!!! Why would I buy a hard drive that I can't use if I change my case? Why would I buy a CD-ROM mounting kit if I have to buy a new one to fit my case? Why would I buy a new motherboard if I have to replace my case to get the RAM installed??? This is a problem throughout the entire computer industry, and not just hardware. Standards need to be set, and followed. Just because a hardware vendor decides that they want to create a different widget, we shouldn't all have to replace other hardware to make it work. Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Let me see, I used to work on UniSys mainframes, and they all had magnetic NDRAM (Non-Destructive RAM), and the problems were :
1. Slower than solid state devices
To change the state of a register you had
to modify the magnetic field
2. Consumes more power. It is more power hungry to
change the state of a magnet than it is to
change the state of a flip-flop
While the idea of Non-Destructive RAM is nice (Power off your machine, power it on and it picks up where it left off), there are much better ways to do this in solid state. You could easily (UniSys did) create a capacitance based NDRAM that isn't as hard to change the state of, and isn't as power hungry.
Just my humble opinion, maybe it makes someone more informed.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Most people who buy video games for their PCs want to go buy them at their local CompUselessA. As far as I can tell, CompUSA and other computer stores aren't carrying the Linux versions of the games. If we can't easily buy them, then you expect that the sales would be poor.
IMHO
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
I posted my initial follow up prior to reading VA response, and I would like to comment on it. I am very impressed with the way VA has owned up to the mistake. It makes me have a great deal more respect for them as a company. Anyone willing to admit that there is a problem, is probably someone who has every intent to fix the problem.
The second point that I would like to make is that not opnly did VA own up to their mistake, but they also defended my own actions in the situation. Although I would never say that I am the ultimate authority on anything, I did have a pretty good handle on the problem. I worked on the problem for at least 1 week prior to calling VA tech support. Initially, I assumed there was a particular patch that VA used to make the kernel compatible with their system, then I wondered if the RAID controller's BIOS was up to date, and finally, I tried running a different set of drive. When all of these failed to repair the problem, it was obvious to me that there was a failed RAID controller (as SCSI drives don't usually fail, and they never fail in pairs).
In truth, some of the blame does lie with me for the server not working, but that does not negate the fact that there were many customer support mistakes made. I wish VA well in their future endeavors, and I hope that they become the company that we all would like to see them become. The Linux community is fast growing, and it is hard for a fast growing company to maintain a consistent level of support when their service requirements are quickly outgrowing their staff's abilities.
My only reason in posting this article was to get an understanding of why over the last five years or so, customer support in general has become so non-existent.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
I don't feel that this is the answer. The reason that Mandrake was formed after Red Hat is that there are different types of hardware, and there are different needs. Windows addresses as many as possible, but they can't address al. Red Hat compiles for i386 instruction sets, whereas Mandrake compiles with the full Pentium instruction set (getting in many cases a %25 performance gain), but if you have a 386, you can't use that distro. I feel that diversity gives the linux community strength.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
To be truthful, from the way that this was posted, it does look this way. In the real world (Read: not what Slashdot posted), I posted a larger article than this. It was about 1.5 pages in total. Where the rest of that text went I'll never know. If you read my follow up you'll see the whole of the information, and if you read VA's post, you'll seen the same story.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc. Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
I originally posted this article with a large amount of additional information. The complete story is as follows.
The server
VA Linux FullOn 2x2
Dual 700MHz PIII
Mylex AcceleRAID 150
4 10000RPM SCSI Drives
Intel 440GX+ Mother Board
Dual Intel EtherExpress Pro 10/100 NICs
The Operating System
Linux Mandrake 7.0 and 7.1 (Both Tried)
ReiserFS andf ext2 (Both Tried)
The first time we recieved the server, it ran the installed Red Hat distro, but due to security concerns and performance issues, we wanted to run Mandrake (Optimized for Pentium processors). The installation went without a hitch every time. After rebooting, the server would have a VFS unable to mount root filesystem, or a kernel oops (aternately). After taking with VA tech support, we reinstalled Their RedHat distro, and had the same results. At their recommendation, I rebuilt the RAID array configuration and reinstalled. During the format of the RAID array one of the drive failure lights began blinking. I shut down the server, reseated the drive, and began the format again. This time two drives failed.
I sent the server back to VA, and after two weeks it was returned to me. When I turned the machine on, there was no OS installed, period. Second, many of the parts we had returned we not returned to us (i.e. the rails, power cord, manuals, faceplate). Again we requested an RMA on the server. The replacement of the server this time took a total of 4 weeks. For an Internet company, that is an eternity when your are trying to beat your competitors to market. I was told many times that the server was being built, and that they had had a bad batch of RAM or a failed burn in. Eventually though, the new server arrived, and this one has worked (Running Mandrake 7.1) since the day after it got here.
VA has offered us on site support, and 3 years parts replacement to attempt to make up for the errors that occured. Although I do not entirely blame the team at VA, I do feel that customer support has been degraded greatly, by MANY companies, and not only VA. We have had similar problems with at least 10 vendors.
This port was not directed entirely at VA, although theirs was the worst example we have had. Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Slashdot Mirror
Doesn't this sort of application really not have much of a real world use? The trying out of a new kernel is okay, but you should be doing that on non-production machines anyhow. And as for cracker protection, chroot works better than usermode linux IMHO. I can see the development testing as being somewhat benficial, but that doesn't really test out the whole kernel does it? The only way to be sure that a kernel truly works is to boot it. These opinions are my own, and not those of my alternate personalities. As such, any litigation should be directed at me sdalj; NOT ME; lkasdlf The dogs told me to do it.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Software and Operating system evaluation is a very complex subject, and by no means is it ever black and white. There are many different ideologies that pervade the field. The international security community is embracing a new system know as Common Criteria(CC). The concept is that you define a set of objectives, and a CC testing facility checks to make sure that your software/OS/Hardware meets those standards. This is much more flexible than the TCSEC(Orange Book) evaluations. This also adds layers of complexity. Which CC eval spec do you need/want, who defines these specs, and how do you get your software tested. Well, the International Information Systems Security Certification Consortium (ISC)^2 for short has many resources for you to find CC specs and CC testing facilities. They also provide a comprehensive training and certification program for people interested in learning about information security. Web Site. The cert is well accepted, but don't think it's going to be a Microsoft MCP exam easy lick - 6 Hours, 250 questions. Hope this is helpful.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
The US was founded on the idea that people from all over the world and become patriated if they offered value to the country. If we deport all of these highly skilled workers, not only will we be hamstringing our Internet economy, but we will be going against some of the basic priciples of our country.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Truthfully, there has to be innovation in the way that cases are being made anyhow. I know that there are many of us in the IT field who run into problems installing X equipment in Y case due to some problem or another. The major concern has to be that vendors follow a set STANDARD (psst... Microsoft, learn this term) that allows all equipment to fit all motherboards/cases/power supplies. I keep running in to cases where I have remove the internal drive bay assembly to fit RAM in with X mother board, or I have to purchase X adapter tray to make Y drive fit Z case. PCs are supposed to be modular!!! Why would I buy a hard drive that I can't use if I change my case? Why would I buy a CD-ROM mounting kit if I have to buy a new one to fit my case? Why would I buy a new motherboard if I have to replace my case to get the RAM installed??? This is a problem throughout the entire computer industry, and not just hardware. Standards need to be set, and followed. Just because a hardware vendor decides that they want to create a different widget, we shouldn't all have to replace other hardware to make it work.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Let me see, I used to work on UniSys mainframes, and they all had magnetic NDRAM (Non-Destructive RAM), and the problems were :
1. Slower than solid state devices
To change the state of a register you had
to modify the magnetic field
2. Consumes more power. It is more power hungry to
change the state of a magnet than it is to
change the state of a flip-flop
While the idea of Non-Destructive RAM is nice (Power off your machine, power it on and it picks up where it left off), there are much better ways to do this in solid state. You could easily (UniSys did) create a capacitance based NDRAM that isn't as hard to change the state of, and isn't as power hungry.
Just my humble opinion, maybe it makes someone more informed.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
Most people who buy video games for their PCs want to go buy them at their local CompUselessA. As far as I can tell, CompUSA and other computer stores aren't carrying the Linux versions of the games. If we can't easily buy them, then you expect that the sales would be poor.
IMHO
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
I posted my initial follow up prior to reading VA response, and I would like to comment on it. I am very impressed with the way VA has owned up to the mistake. It makes me have a great deal more respect for them as a company. Anyone willing to admit that there is a problem, is probably someone who has every intent to fix the problem.
The second point that I would like to make is that not opnly did VA own up to their mistake, but they also defended my own actions in the situation. Although I would never say that I am the ultimate authority on anything, I did have a pretty good handle on the problem. I worked on the problem for at least 1 week prior to calling VA tech support. Initially, I assumed there was a particular patch that VA used to make the kernel compatible with their system, then I wondered if the RAID controller's BIOS was up to date, and finally, I tried running a different set of drive. When all of these failed to repair the problem, it was obvious to me that there was a failed RAID controller (as SCSI drives don't usually fail, and they never fail in pairs).
In truth, some of the blame does lie with me for the server not working, but that does not negate the fact that there were many customer support mistakes made. I wish VA well in their future endeavors, and I hope that they become the company that we all would like to see them become. The Linux community is fast growing, and it is hard for a fast growing company to maintain a consistent level of support when their service requirements are quickly outgrowing their staff's abilities.
My only reason in posting this article was to get an understanding of why over the last five years or so, customer support in general has become so non-existent.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
I don't feel that this is the answer. The reason that Mandrake was formed after Red Hat is that there are different types of hardware, and there are different needs. Windows addresses as many as possible, but they can't address al. Red Hat compiles for i386 instruction sets, whereas Mandrake compiles with the full Pentium instruction set (getting in many cases a %25 performance gain), but if you have a 386, you can't use that distro. I feel that diversity gives the linux community strength.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
To be truthful, from the way that this was posted, it does look this way. In the real world (Read: not what Slashdot posted), I posted a larger article than this. It was about 1.5 pages in total. Where the rest of that text went I'll never know. If you read my follow up you'll see the whole of the information, and if you read VA's post, you'll seen the same story. Deven Phillips, CISSP Network Architect Viata Online, Inc.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.
I originally posted this article with a large amount of additional information. The complete story is as follows. The server VA Linux FullOn 2x2 Dual 700MHz PIII Mylex AcceleRAID 150 4 10000RPM SCSI Drives Intel 440GX+ Mother Board Dual Intel EtherExpress Pro 10/100 NICs The Operating System Linux Mandrake 7.0 and 7.1 (Both Tried) ReiserFS andf ext2 (Both Tried) The first time we recieved the server, it ran the installed Red Hat distro, but due to security concerns and performance issues, we wanted to run Mandrake (Optimized for Pentium processors). The installation went without a hitch every time. After rebooting, the server would have a VFS unable to mount root filesystem, or a kernel oops (aternately). After taking with VA tech support, we reinstalled Their RedHat distro, and had the same results. At their recommendation, I rebuilt the RAID array configuration and reinstalled. During the format of the RAID array one of the drive failure lights began blinking. I shut down the server, reseated the drive, and began the format again. This time two drives failed. I sent the server back to VA, and after two weeks it was returned to me. When I turned the machine on, there was no OS installed, period. Second, many of the parts we had returned we not returned to us (i.e. the rails, power cord, manuals, faceplate). Again we requested an RMA on the server. The replacement of the server this time took a total of 4 weeks. For an Internet company, that is an eternity when your are trying to beat your competitors to market. I was told many times that the server was being built, and that they had had a bad batch of RAM or a failed burn in. Eventually though, the new server arrived, and this one has worked (Running Mandrake 7.1) since the day after it got here. VA has offered us on site support, and 3 years parts replacement to attempt to make up for the errors that occured. Although I do not entirely blame the team at VA, I do feel that customer support has been degraded greatly, by MANY companies, and not only VA. We have had similar problems with at least 10 vendors. This port was not directed entirely at VA, although theirs was the worst example we have had.
Deven Phillips, CISSP
Network Architect
Viata Online, Inc.