ldm@texhrc.UUCP (Lyle Meier) (04/07/91)
We have a decssystem 5840 running ultrix 4.1 and a sgi 4d/25 and 43/310 running 3.3.2. It is possible to mount the ultrix file systems on the sgi using automount, but it is not possible to mount the sgi file systems on the ultrix machine. However if a hard mount is done it is possible to perform this task. When the automount is tried from ultrix the message permission denied appears along with the message no such file. Any ideas where the problem may lie. The SGI hotline has been no help, I recognize that this problem lies sort of between vendors. With suns the sgi file system automount ok.
mikeew@cascade.ens.tek.com (Michael Ewan) (04/08/91)
I solved a similar problem by using amd. Get it from gatekeeper.dec.com. Mike
shc@lhdsy1.chevron.com (S.H. Couturie) (04/08/91)
In article <572@texhrc.UUCP> ldm@texhrc.UUCP (Lyle Meier) writes: > >We have a decssystem 5840 running ultrix 4.1 and a sgi 4d/25 and 43/310 >running 3.3.2. It is possible to mount the ultrix file systems on the >sgi using automount, but it is not possible to mount the sgi file >systems on the ultrix machine. However if a hard mount is done >it is possible to perform this task. >When the automount is tried from ultrix the message permission denied >appears along with the message no such file. > >Any ideas where the problem may lie. The SGI hotline has been no help, >I recognize that this problem lies sort of between vendors. >With suns the sgi file system automount ok. The Ultrix automounter does not issue the mount request properly to a reserved port; it is as if the mount request is coming from a non-root user. SGI IRIX restricts (properly, IMHO) mounts to the super-user, and accordingly normally only accepts NFS mount requests from clients on reserved ports. There is a way around this Ultrix problem: on the Iris, in /usr/etc/inetd.conf, change the "mountd" lines to end in "mountd -n" : mountd/1 stream rpc/tcp wait root /usr/etc/rpc.mountd mountd -n mountd/1 dgram rpc/udp wait root /usr/etc/rpc.mountd mountd -n sgi_mountd/1 stream rpc/tcp wait root /usr/etc/rpc.mountd mountd -n sgi_mountd/1 dgram rpc/udp wait root /usr/etc/rpc.mountd mountd -n (I changed them all, for certainty; probably someone at SGI can say which ONE needs to be changed.) This problem also occurs when mounts are issued from Apollo systems. And, IBM's AIX acts the same as SGI IRIX does here, with a similar solution (although in AIX the mountd is run continuously as a daemon). Give this a try, and good luck. Steve Couturie ---- Steve Couturie Voice: (213) 694-9332 Chevron Oil Field Research Co. FAX: (213) 694-7228 P.O. Box 446 Internet: shc@chevron.com La Habra, CA 90633-0446 UUCP: ...!uunet!lhdsy1!shc