As far as I can tell, samba follows symlinks very nicely. (mixture of home directories in tomcat's webapps and having available space in the "wrong" directories.)
I think it's the server that would get stuck, not the attacker.:-(
It's not a buffer overflow in just VB. It's a buffer overflow in VB, in C, in C++, in C# or any language on Windows that can call the API. Should be a lot of them.
PHP is simple, direct, and very effective. Probably does not scale well to enterprise levels.
Java looks like it is targeted at the mainframes of 5 years into the future with an overall complexity that would make SAP look simple. Somehow I think the real target of Java is the masses of COBOL programs in banks, insurance companies, etc.
Actually Java and PHP work pretty well together. We're using Apache/Tomcat with jsp and php in same directories.
Try redhat.com/errata
In addition to links to Errata for 7.1 going back to 4.0,
Notable Security Exploits
Red Hat Linux users who have applied all Red Hat security updates are usually not vulnerable to worms and other security exploits. Click on the links below to read about each recent exploit and what you can do to prevent being affected.
The Adore Worm
Discovered April 3, 2001
The Lion Worm
Discovered March 23, 2001
Bind Exploit
Discovered January 29, 2001
The Ramen Noodle Worm
Discovered January, 2001
http://openbsd.com/errata.html
even better organized
My sympathies to anyone using Windows or Dual-Booting, but your answer is the only way to get rid of it.
Remember Melissa? Melissa was nice. This latest whatever will not be the last..NET anyone?
The advertising revenue after things return to normal and people still want to get their daily dose of CNN. The Gulf War boosted CNN enormously, not just during the Gulf War.
>>Conduct remedies can't effectively be enforced in this case, therefore they are equivalent to nothing.
Civil suits. Lost profits. Messy and expensive ( for Microsoft;)
It strikes me that there are two main differences between bridges and software.
One, the design and components of the bridge are public knowledge.
Two, when the bridge does fall down, you have a lot of people finding out what went wrong.
As for complexity, I would imagine that any bridge is actually much more complicated than any computer program. (subatomic forces vs machine state)
Or something. It's the normal DOS/Windows C: drive. It's not the "second" anything. The first or only drive. The first or only partition on that drive. It is useful to remember what that drive/partition is called on a system that is capable of handling many partitions on many drives.
There is very little arbitrary about hda1. "a" as in "abcdefg...". Kindergarten stuff. "1" as in "123456...". Again, kindergarten stuff. "hd" as in "hard drive". "sd" for SCSI drive. How is "drdrv" or "hdd" any better than "hd"? Or do you prefer "scsi(0)disk(0)rdisk(0)partition(1)"?
Is C: any better or more intuitive?
First partition of the master drive of the primary IDE controller?
Now just try to maintain sanity with a SCSI controller or two.
For more fun, add a Promise ATA controller to the mix.
Re:In my experience, you are both right.
on
Linux Office Suites
·
· Score: 1
In my experience, you are both right. It is possible to recover from pulling the plug in Windows. It's not only possible, but a lot more likely.
Microsoft products seem to me to be built with the idea that they need to be buggy so that there will be a reason to upgrade. More likely it's the "single-user" paradigm. Further, each app thinks it's the most important thing in the world, and everything should be arranged for its benefit. This leads to pervasive assumptions that in the real world have bizarre consequences. However, at this point, Microsoft is hurting for expansion possibilities and is likely to try just about anything to try to prolong its position.
First of all, the assertion that the company would HAVE to move to per-seat licensing when they moved to separate file, print and mail servers is just wrong. 2000 concurrent users are still 2000 concurrent users, whether they are connected to one server or three. Sounds good to me, but will Microsoft buy it?
For self-preservation, install two (2) systems and using the second, copy WINNT\SYSTEM32\CONFIG to something like WINNT\SYSTEM32\CONFIG-BAK. Saved my hide a couple of times;)
The point is that for stability reasons, it takes a 6 server NT cluster to replace 1 Linux/FreeBSD box. 3 of the servers are there because even after splitting the server functions to different servers, the live backup servers are required to provide adequate reliability. What really makes this situation rediculous is that Linux is not (yet) "Enterprise-Ready".
No, impressed to see it running. Depressed trying to get it running, (or trying to keep it running;-)
I wonder if this (sendmail on big iron) is able to handle an enterprise-wide outbreak of the next outlook virus/worm/whatever.
Slashdot Mirror
Hmmm, interesting.
What's a poor worm to do?
OT, but why does OpenBSD have command recall and tab completion in ftp and not in sh?
So, Microsoft is unconstitutional?
As far as I can tell, samba follows symlinks very nicely. (mixture of home directories in tomcat's webapps and having available space in the "wrong" directories.) :-(
I think it's the server that would get stuck, not the attacker.
It's not a buffer overflow in just VB. It's a buffer overflow in VB, in C, in C++, in C# or any language on Windows that can call the API. Should be a lot of them.
PHP is simple, direct, and very effective. Probably does not scale well to enterprise levels.
Java looks like it is targeted at the mainframes of 5 years into the future with an overall complexity that would make SAP look simple. Somehow I think the real target of Java is the masses of COBOL programs in banks, insurance companies, etc.
Actually Java and PHP work pretty well together. We're using Apache/Tomcat with jsp and php in same directories.
I'd say yes.
l
Try redhat.com/errata
In addition to links to Errata for 7.1 going back to 4.0,
Notable Security Exploits
Red Hat Linux users who have applied all Red Hat security updates are usually not vulnerable to worms and other security exploits. Click on the links below to read about each recent exploit and what you can do to prevent being affected.
The Adore Worm
Discovered April 3, 2001
The Lion Worm
Discovered March 23, 2001
Bind Exploit
Discovered January 29, 2001
The Ramen Noodle Worm
Discovered January, 2001
http://openbsd.com/errata.html
even better organized
http://www.freebsd.org/releases/4.3R/errata.htm
LOL.
RedHat 8.2 will have enough compatability and Sandbox/Jail capabilities to not only run it, but to run it with impunity.
My sympathies to anyone using Windows or Dual-Booting, but your answer is the only way to get rid of it. .NET anyone?
Remember Melissa? Melissa was nice. This latest whatever will not be the last.
It's Microsoft's idea of security.
Speaking of Imperial, cars get more miles per gallon in Canada than they do in the US.
The advertising revenue after things return to normal and people still want to get their daily dose of CNN. The Gulf War boosted CNN enormously, not just during the Gulf War.
to be followed by Windows x P U
Exactly.
How to live with the unbearable.
You can laugh or cry. Laughing is better.
>>Conduct remedies can't effectively be enforced in this case, therefore they are equivalent to nothing. ;)
Civil suits. Lost profits. Messy and expensive ( for Microsoft
It strikes me that there are two main differences between bridges and software.
One, the design and components of the bridge are public knowledge.
Two, when the bridge does fall down, you have a lot of people finding out what went wrong.
As for complexity, I would imagine that any bridge is actually much more complicated than any computer program. (subatomic forces vs machine state)
Now that sounds like the basis for a real mean virus/worm/whatever. Don't format the hard drive, just mess with the product activation.
Or something. It's the normal DOS/Windows C: drive. It's not the "second" anything. The first or only drive. The first or only partition on that drive. It is useful to remember what that drive/partition is called on a system that is capable of handling many partitions on many drives.
There is very little arbitrary about hda1. "a" as in "abcdefg...". Kindergarten stuff. "1" as in "123456...". Again, kindergarten stuff. "hd" as in "hard drive". "sd" for SCSI drive. How is "drdrv" or "hdd" any better than "hd"? Or do you prefer "scsi(0)disk(0)rdisk(0)partition(1)"?
Is C: any better or more intuitive?
First partition of the master drive of the primary IDE controller?
Now just try to maintain sanity with a SCSI controller or two.
For more fun, add a Promise ATA controller to the mix.
In my experience, you are both right. It is possible to recover from pulling the plug in Windows.
It's not only possible, but a lot more likely.
Microsoft products seem to me to be built with the idea that they need to be buggy so that there will be a reason to upgrade.
More likely it's the "single-user" paradigm. Further, each app thinks it's the most important thing in the world, and everything should be arranged for its benefit. This leads to pervasive assumptions that in the real world have bizarre consequences. However, at this point, Microsoft is hurting for expansion possibilities and is likely to try just about anything to try to prolong its position.
This is Microsoft. The software is integrated. A bad driver on NT Workstation causes a crash on NT Server.
First of all, the assertion that the company would HAVE to move to per-seat licensing when they moved to separate file, print and mail servers is just wrong. 2000 concurrent users are still 2000 concurrent users, whether they are connected to one server or three.
Sounds good to me, but will Microsoft buy it?
For self-preservation, install two (2) systems and using the second, copy WINNT\SYSTEM32\CONFIG to something like WINNT\SYSTEM32\CONFIG-BAK. Saved my hide a couple of times ;)
Yeah, it reads like a fairy tale. I imagine the reality is even more lopsided.
The point is that for stability reasons, it takes a 6 server NT cluster to replace 1 Linux/FreeBSD box. 3 of the servers are there because even after splitting the server functions to different servers, the live backup servers are required to provide adequate reliability. What really makes this situation rediculous is that Linux is not (yet) "Enterprise-Ready".
No, impressed to see it running. Depressed trying to get it running, (or trying to keep it running ;-)
I wonder if this (sendmail on big iron) is able to handle an enterprise-wide outbreak of the next outlook virus/worm/whatever.
Some things are too important to be taken seriously. /. seems to me more like a soap opera, and just as addictive.