Hi all,
sorry it became rather lenghty...:
Why does my NIS not work or do I not understand how it should?
I installed NIs on several DU systems
Node
6 Master DU V4.0D 878 (from uname -a)
2 slave V4.0 386
3 client V4.0b 564
4 client V4.0D 564.32
5 client V4.0D 878
/var/yp/src contains passwd,group,hosts
After some tries NIS started properly on all systems
I tested to change my password
yppasswd on 6(master) changed from p1 to p2
but:
login on 5 only with old pwd p1
login from 6 to 6 only with old pwd p1
(and on all others)
now:
yppasswd on 6 only to change with p2 to p3
login to 3,5 still with p1
yppasswd on 3,5 from p3 to p4 (results in mismatch)
from p1 to p4 (results couldn't change)
yppasswd on 6 (master) changes only from p3 to p4
>From this I conclude my NiS password has nothing to do with my
login password. Maybe I don't understand but this was the reason to
have the same password on all nodes (like in a VMS Cluster). (I did
this manually until now by copying regularly the passwd files from
one node to the others)
On the different nodes are running
6 master rpc.yppasswd /var/yp/src/passwd
ypserv -a d
ypbind -s -S domain,master-node
2 slave ypbind -s -S domain,slave-node
ypserv
5 client ypbind -s -S domain,master-node
(I used for the names of all nodes the full IP address e.g.
node6.internet-domain. The domain is a simple name "linax" but the
same on all nodes)
I tried to use ypxfr with the passwd maps (because I don't know how
the update is done between the nodes) on:
2 slave: Message: Information is not newer
3 client Message: can't write
(this seems reasonable because the directory
/var/yp/domain doesn't exist)
I even rebooted all systems but got the same results afterwards as
above.
What I am doing wrong? Or what I am not understanding?
Thank you for every hint.
Best regards
Otto
PS: Assumed I get the whole thing running. What do I have to do if
I make an adduser on the master
- copy passwd to /var/yp/src ?
- or create a link /var/yp/src/passwd -> /etc/passwd ?
- makedbm ?
And if are there somewhere template scripts because this might be
a very common procedure?
-------------------------------------------------------------------------
| Dr. Otto Titze, Kernphysik TU, Schlossgartenstr. 9, D-64289 Darmstadt |
| titze_at_ikp.tu-darmstadt.de Tel: +49(6151)16-2916,FAX:16-4321|
-------------------------------------------------------------------------
Received on Wed Dec 09 1998 - 18:53:16 NZDT