On Feb 25, 11:10am, i18rabbit@cwazy.co.uk (i18rabbit@cwazy.co.uk) wrote:
-- Subject: Re: bin/32903: utmp remains empty

| The following reply was made to PR bin/32903; it has been noted by GNATS.
| 
| From: i18rabbit@cwazy.co.uk
| To: gnats-bugs@netbsd.org
| Cc: 
| Subject: Re: bin/32903: utmp remains empty
| Date: Sat, 25 Feb 2006 11:08:40 GMT
| 
|  >  >  utmp entries are created by ssh logins and reboots/shutdowns,
|  >  >  but not for *any* console or telnet logins.  this is the problem.
|  > 
|  >  Yes, but it's not a problem on a properly-installed NetBSD 3.0 system.
|  > 
|  >  >  because i maintain several BSD systems, and i have
|  >  >  a stripped custom /etc directory that i use for
|  >  >  all of them; for ease of set-up and maintenance.
|  > 
|  >  That strikes me as a bad idea. You should've said that originally... In
|  >  that case, all I can say is that you're on your own if your custom
|  >  modifications cause things to break.
|  
|  my "custom modifications" worked with the NetBSD 2 series.
|  besides - since when aren't "custom modifications" allowed?
|  bringing up a UNIX system is a fairly generic routine.
|  
|  to the best of my memory (7 years or so), the kernel spawns
|  init, which spawns getty's, which spawn login's - which write
|  to utmp/wtmp files; all which are running as root.
|  
|  my "custom modifications" (who does NOT have any of those?)
|  shouldn't be relevant.  what would be relevant is for someone
|  with knowledge of login(1) to state a test that can be done
|  to resolve the problem of login(1) not writing to utmp/wtmp
|  files.  it shouldn't be too complicated to resolve this.

Do you have /etc/pam.d?

christos