Subject: Re: NetBSD-release-2-0 build fails
To: Ian Zagorskih <ianzag@megasignal.com>
From: Steven M. Bellovin <smb@research.att.com>
List: current-users
Date: 06/02/2004 10:18:01
In message <200406022118.20812.ianzag@megasignal.com>, Ian Zagorskih writ=
es:
>=D0=92 =D1=81=D0=BE=D0=BE=D0=B1=D1=89=D0=B5=D0=BD=D0=B8=D0=B8 =D0=BE=D1=82=
 2 =D0=98=D1=8E=D0=BD=D1=8C 2004 20:59 Steven M. Bellovin =D0=BD=D0=B0=D0=
=BF=D0=B8=D1=81=D0=B0=D0=BB(a):
>> In message <200406021946.51375.ianzag@megasignal.com>, Ian Zagorskih w=
rites:
>> >$ uname -a
>> >NetBSD IANZAG 2.0_BETA NetBSD 2.0_BETA (IANZAG) #2: Wed May 26 13:01:=
57
>> > NOVST 2004  ianzag@IANZAG:/usr/src/sys/arch/i386/compile/IANZAG i386=

>> >
>> >Completely updated from today's CVS
>>
>> That's odd -- I did a 'build.sh release' yesterday, and a 'cvs update'=

>> just now showed no changes.
>>
>
>I tried to build release yesterday too but failed on the same sendmail's=
 =

>place :(
>
># ident /usr/src/gnu/usr.sbin/sendmail/Makefile
>/usr/src/gnu/usr.sbin/sendmail/Makefile:
>     $NetBSD: Makefile,v 1.6 2003/07/04 04:52:59 atatat Exp $
>
>AFAIU there should be .cf files that install expects to see in /etc/mail=
/ but =

>they are missing in my DESTDIR:
>
># ls -l /usr/build/dest/i386/etc/mail
>total 16
>-rw-r--r--  1 root  wheel  1346 Jun  6  2002 aliases
>-r--r--r--  1 root  wheel  5648 Mar 26 01:14 helpfile
>-rw-r--r--  1 root  wheel     0 Jun  2 21:06 local-host-names

Very strange!

b220$ cd /usr/BUILD/netbsd/etc/mail
b221$ ls -l
total 105
-rw-r--r--  1 root  wheel   1346 Jun  2 02:09 aliases
-r--r--r--  1 root  wheel   5648 Jun  1 23:59 helpfile
-rw-r--r--  1 root  wheel      0 Jun  2 02:09 local-host-names
-r--r--r--  1 root  wheel  59062 Jun  2 02:09 sendmail.cf
-r--r--r--  1 root  wheel  39537 Jun  2 02:09 submit.cf

I'll ask two semi-random questions:  did you do your build with a clean =

obj and destination tree, and were you using NFS or read-only overlays =

or some such?  Both are frequently mentioned in the context of build =

failures (but I have no idea if either is involved here).

		--Steve Bellovin, http://www.research.att.com/~smb