[11] in Zephyr_Comments
zephyr starting up on 6.0
daemon@ATHENA.MIT.EDU (qjb@ATHENA.MIT.EDU)
Tue Apr 12 14:37:13 1988
From: <qjb@ATHENA.MIT.EDU>
To: rt-testers@ATHENA.MIT.EDU, zephyr-comments@ATHENA.MIT.EDU
Cc: henry@garp.mit.edu, qjb@ATHENA.MIT.EDU
Date: Tue, 12 Apr 88 14:36:25 EDT
A while ago, I made a few comments about zephyr not behaving properly
when I logged into my private worksation remotely either through a
dialup server or from another workstation on campus. The windowgram
client (zwgc) was not starting up at all, and I was not being zlogged
in to my workstation.
The problem turns out to be a simple one: the "real" zlogin and
zlogout are not on the root partitions of worksatations; therefore,
since I often login to my workstation when it is in a deactivated
state and activate it by calling /etc/athena/activate from my .login
sequence, zlogin and zwgc do not exist until login has already tried
to use them. The question of whether zlogin and zwgc should be on the
root partition of each workstation is obviously debatable, but if it
is decided that they should not, at least
LOGIN SHOULD SAY WHEN IT CANNOT FIND THEM.
Something like
login: unable to start Windowgram Client
login: unable to register you with Zephyr
would be a lot more friendly then the confusion that occurs when things are
not working the way they should be.
Jay Berkenbilt
Athena "watchmaker" and olc volunteer
