HNHacker News
TopNewBestAskShowJobs

KonradKlause

636 karma · joined May 12, 2011

submissionscomments
KonradKlause··on One Million Concurrent TCP connections
conntrack is not netstat. I requires netfilter's connection tracking feature.

Using conntrack on a 1MC system will waste even more kernel memory!

KonradKlause··on Overflows in SafeInt
Like any other spec. :-)
KonradKlause··on Overflows in SafeInt
Some background on integer overflows:

http://www.fefe.de/intof.html

KonradKlause··on Cracking OSX Lion passwords
No, OpenSUSE for example is using bcrypt.

See: http://www.openwall.com/crypt/

KonradKlause··on Cracking OSX Lion passwords
Thanks for the clarification!
KonradKlause··on Cracking OSX Lion passwords
TL;DR:

There is no need to crack the password. You (as non-root user) can just reset the currently logged in user's password by calling:

dscl localhost -passwd /Search/Users/bob

KonradKlause··on AeroFS (YC S10): It’s Like Dropbox, But It Runs Inside Your Firewall
WTF? Why is this being down-voted?
KonradKlause··on AeroFS (YC S10): It’s Like Dropbox, But It Runs Inside Your Firewall
Ok, no source (jet).

Please vote: http://vote.aerofs.com/forums/67721-feature-requests/suggest...

KonradKlause··on AeroFS (YC S10): It’s Like Dropbox, But It Runs Inside Your Firewall
Where is the source code?
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
Sorry, you're just wrong.

Have you ever called a system call by hand using assembly?

Have you ever looked at glibc's source?

Habe you ever looked at the kernel's close() syscall?

KonradKlause··on Kernel module for advanced rickrolling replaces open() call
open(2) is the wrapper within the libc. There is no open(3) on Linux.
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
It's open(2). Yes, here on my x86_64 it can be hooked. But I've seen lots of setups where this was not doable. As I said it depends how the libc's wrapper is written.
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
With LD_PRELOAD you can hook any _function_ call. A syscall is something different. It depends how the syscall wrapper within the libc is coded. Some can be hooked...
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
Can you name such a module?
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
Using LD_PRELOAD you cannot hook a syscall like open().
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
> It uses strncpy_from_user to do this and uses PATH_MAX as the maximum length. So even the linux kernel does rely on the fact that there is a null terminator in the string.

But the kernel does not call strlen() on the user-supplied string...

KonradKlause··on Kernel module for advanced rickrolling replaces open() call
Sure.

Suppose your admin is a funny guy and loads this kernel module at April 1st. I (Jonny non-root user) can no longer listen to my favorite mp3, because the kernel module changes the path all the time. The admin loughs and has fun.

But then later I call open() with some fancy arguments and take over the whole machine (or at least crash it). This is the part where I lough.

KonradKlause··on Kernel module for advanced rickrolling replaces open() call
Manipulating the cr0 register is _very_ kernel specific.
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
And you are obviously missing that modern operating systems have more than just one user. :-)
KonradKlause··on Kernel module for advanced rickrolling replaces open() call
BTW: write_cr0(read_cr0() | 0x10000)

is a absolute no-go. It's a nice and funny kernel module but it dooms the kernel's security.

KonradKlause··on Kernel module for advanced rickrolling replaces open() call
1. A kernel module is more 1337 ;)

2. Using FUSE you would have to guaranty that your fs overlaps all other mounted filesystems.

KonradKlause··on Kernel module for advanced rickrolling replaces open() call
Bullshit.

Anyone who's using open() can kill the kernel after loading this module.

KonradKlause··on Kernel module for advanced rickrolling replaces open() call
p = (char *)(path + strlen(path) - 4);

This is a very bad idea as path is user-supplied and has to be treated as malicious. An attacker can omit the string-terminator...

KonradKlause··on Kernel.org has been hacked (see site news)
Of course not.

As one of the users@kernel.org members I can tell you that the kernel.org admins are very competent.

KonradKlause··on Kernel.org has been hacked (see site news)
Currently all files get verified. The kernel.org admins have digital signatures of all files...
KonradKlause··on Ddate(1) disabled by default in future util-linux
Damn lkml.org is down:

http://marc.info/?l=linux-kernel&m=131462065108921&w...

KonradKlause··on [dead]
https://bitcointalk.org/index.php?PHPSESSID=0f35e6daf7c9d42b...

made my day. :-)

KonradKlause··on Oracle Buys Ksplice
Time to fork it!

http://knoppix.mirrors.tds.net/pub/linux/frugalware/frugalwa...

KonradKlause··on Every programmer should read the source to abort() at some point in their life.
ACK
KonradKlause··on Every programmer should read the source to abort() at some point in their life.
very true. but glibc's abort() is IMHO ok.
← PreviousPage 2 of 4Next →