Re: NIS issues
In article <cistron.20000402145828.A20491@cs.mcgill.ca>,
Nick Cabatoff <ncc@cs.mcgill.ca> wrote:
>I keep getting mail on some slink machines from cron jobs, e.g. the
>"test -f /proc/modules && /sbin/rmmod -a" one. They contain the
>single line:
>
> yp_all: clnt_call: RPC: Unable to receive; errno = Connection reset by
>peer
>
>Anyone know what this signifies,
NIS temporarily failed to talk to the NIS server, then it printed an
error to stderr (it's the NIS routines in glibc that do this -
bad, I think, but oh well). Usually it's just an informational message
as it automatically switches servers if one is down or not
answering immidiately.
>how NIS gets involved in such a
>simple cronjob
It probably did something that uses NIS, i.e. getpwnam() or getpwuid()
>and what I can do about it?
Usually it is a very good idea to list the NIS servers directly
in /etc/yp.conf instead of relying on broadcasts (the default if
/etc/yp.conf is empty), it makes the NIS client stuff a bit more
reliable. /etc/init.d/nis stop and start after that.
Mike.
--
Windows never had any potential for soundness or beauty. If you decide to
build a motorcycle, and you start with a bathtub, no good will ever come of it.
-- Anonymous Coward
Reply to: