From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)
Description of problem:
The environment variable REMOTEHOST is not always getting set when using
rsh/rlogin to remotely log into a RH v7.2 machine. This is a strange one,
since there is no discernable pattern of why this is failing. I can rsh/rlogin
into a RH v7.2 machine and sometimes REMOTEHOST does not get set. If I reboot
the machine that is the destination, then it appears to work for a bit of
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1.rsh/rlogin into v7.2 machine
2. env|grep REMOTE
If the remote login has successfully set $REMOTEHOST, then step 2 should
produce something along the lines of:
Within our current production environment, we commonly log into remote
machines. During any type of login, we test to see if REMOTEHOST is defined.
If REMOTEHOST is not defined, we launch an application that changes the color
profile of the machine's graphics card. During a remote login the application
that changes the color profile gets launched and core dumps if REMOTEHOST is
not defined. This extremely annoying and is preventing us from putting in any
other logic that would allow us to tailor our logins.
This only appears to be a problem on the first windowed-shell opened on the
machine, or at least that's how it's behaving today. I'm a little confused
though, is the first windowed shell pts/0 or pts/1? The odd thing is the
problem seems to come and go and we haven't been able to specifically put a
finger on what exactly is causing the problem.
Have you been able to reproduce it consistently in the meantime? A bug that not
really reproducible is very difficult to fix.
Read ya, Phil
I wasn't able to reproduce this bug recently. It seems to be fixed in current