Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!hsdndev!husc6!zariski!primer From: primer@math.harvard.edu (Jeremy Primer) Newsgroups: comp.unix.admin Subject: Re: NIS question Message-ID: Date: 10 Dec 90 04:29:49 GMT References: <1990Dec4.182751.10341@amd.com> <1990Dec7.022705.8901@wdl1.wdl.loral.com> Sender: news@husc6.harvard.edu Organization: Harvard University Dept. of Mathematics Lines: 36 In article <1990Dec7.022705.8901@wdl1.wdl.loral.com> rob@wdl50.wdl.fac.com (Rob Riepel) writes: >I'm currently taking the SunOS system admin class and we were told that there >is a bug in rpc.yppasswdd that prevents it from doing the automatic update of >the passwd map. I don't know whether Sun has volunteered the explanation for this bug, so for the record, here is what I observed while trying to get yppasswdd to work. Our rc.local line prior to SunOS 4.1 was 'rpc.yypasswdd /etc/passwd -m'. I figured the problem might be something amiss in the make in /var/yp. I observed that /var/yp/Makefile builds target "k" instead of target "all" by default, and that target "k" simply reruns make with the -k option, telling make to attempt successive targets even if some fail. Thus inspired, I tried yppasswdd as: rpc.yypasswdd /etc/passwd -m -k passwd and it worked. The question remained: what were the errors which -k convinced make to ignore? yppasswdd had to be invoking make incorrectly. So I changed my NIS password and then immediately ran 'ps ax | grep passwd' in another window and saw: 64 ? S 0:00 rpc.yppasswdd /etc/passwd -m -k passwd 28027 ? S 0:00 rpc.yppasswdd /etc/passwd -m -k passwd 28028 ? S 0:00 sh -c make rpc.yppasswdd /etc/passwd -m -k passwd 28029 ? S 0:00 make rpc.yppasswdd /etc/passwd -m -k passwd 28043 pd S 0:00 grep passwd The -k option convinces make to work even though it has been passed some inappropriate arguments. It seems that there are other workarounds which cause rpc.yppasswdd to invoke make properly. -- Jeremy Primer, Department of Mathematics, 1 Oxford Street, Cambridge MA 02138 primer@math.harvard.edu ...!harvard!zariski!primer primer@huma1.bitnet