Re: openssh problem after going from 5.2.1 to 5.3-beta7

2004-10-12 Thread Michael Ray
On Mon, 11 Oct 2004 11:48:06 -0700, you wrote: I have a HP DL380g3 I was running 5.2.1-REL on (I think it was on -p9). I did a source upgrade to 5.3-beta7, including mergemaster -p, followed by mergemaster which did upgrade /etc/ssh/sshd_config. When I am now trying to connect from SecureCRT, I

Why present different size and md5 hash between the compiled code and FreeBSD's build-in binary ?

2004-10-12 Thread Jerry
Hi~ This is my first time to post question here, and also I am a newbie on FreeBSD. A confusing thing I discover is: Why the binary code I compiled(for example,telnet command)compare with the original one is diffirent, either size or md5 hash value, What's going on? Are the source files differ

Re: Why present different size and md5 hash between the compiled code and FreeBSD's build-in binary ?

2004-10-12 Thread Andreas Kohn
Jerry wrote: Hi~ This is my first time to post question here, and also I am a newbie on FreeBSD. A confusing thing I discover is: Why the binary code I compiled(for example,telnet command)compare with the original one is diffirent, either size or md5 hash value, What's going on? Are the source

Re: Why present different size and md5 hash between the compiled code and FreeBSD's build-in binary ?

2004-10-12 Thread Andreas Kohn
Jerry wrote: --- Andreas Kohn [EMAIL PROTECTED] wrote Hi, you might be using different optimization flags than the release building cluster. Also make sure that you have really the correct sources, and not a newer version from CVS. Hi~ Andreas, I just used make command with default

Why present different size and md5 hash between the compiled code and original binary ?

2004-10-12 Thread Jerry
Hi~ All This is my first time to post question here, and also I am a newbie on FreeBSD. A confusing thing I discover is: Why the binary code I compiled(for example,telnet command)compare with the original one is diffirent, either size or md5 hash , What's going on? Are the source files differ

Re: Why present different size and md5 hash between the compiled code and FreeBSD's build-in binary ?

2004-10-12 Thread Max Laier
On Tuesday 12 October 2004 13:11, Andreas Kohn wrote: Jerry wrote: --- Andreas Kohn [EMAIL PROTECTED] wrote Hi, you might be using different optimization flags than the release building cluster. Also make sure that you have really the correct sources, and not a newer version from

Re: Why present different size and md5 hash between the compiled code and FreeBSD's build-in binary ?

2004-10-12 Thread Joerg Sonnenberger
On Tue, Oct 12, 2004 at 01:11:09PM +0200, Andreas Kohn wrote: if your source files are *exactly* the same versions used as on the building cluster, and you have *exactly* the same compilation options, it would still be possible to have different binaries. For example if the files included some

malloc calls and ioctl calls to soundcard cause segfault

2004-10-12 Thread Shawn Webb
I have stumbled upon a local DoS (non-kernel) while writing a VoIP app for FreeBSD. The DoS exists when two ioctl calls (or less/more?) are followed by a malloc call to malloc a pointer in global scope which is then followed by two more (or less/more?) ioctl calls. The result is a stack smash,

Why present different size and md5 hash between the compiled

2004-10-12 Thread Jerry
Thank you all for explaining so patiently, I have somewhat of understanding with this issue ~ very appreciated all who commented. Jerry - Yahoo! 100MB http://mail.yahoo.com.tw/ ___ [EMAIL