[comp.protocols.nfs] TCP/IP and NFS on ISC

geoff@hinode.East.Sun.COM (Geoff Arnold @ Sun BOS - R.H. coast near the top) (08/31/89)

In article <899@orbit.UUCP> jeb@pnet51.cts.com (James Borchart) writes:
>[Deleted: various problems with ISC Unix TCP/IP and NFS]
>
>        3: SUN's PC-NFS, running on a PC connected by ethernet to the
>           ISC server, gets "authentication failure" no matter what is typed
>           for the username and password, or if the user needs a password.
> 

I don't have a copy of the ISC stuff running here (lame excuse: I should
order one), but assuming that their PCNFSD is derived from the standard
source you should be able to start it with a "-d" option to trace
each transaction. Of course this assumes that PCNFSD is running:
to check this, run the following on the PC:

	rpcinfo -u <servername> 150001

>[...]                            Also, does anyone know how to modify
>the printer timeout time for PC-NFS, 5 minutes is a little extreme.
>
Fixed in 3.0.1. If you don't have a copy, contact Sun's Customer
Support Division for upgrade info. (Or buy another copy of PC-NFS
and use it to upgrade existing nodes as well....)


Geoff Arnold,                              Internet: geoff@East.Sun.COM
PCDS Group, Sun Microsystems Inc.
---------------------------------------------------------------------------
Most bizarre title of the week: "LAN Standards - Do you Need Them?" (in BYTE)

sauer@dell.dell.com (Charlie Sauer) (09/01/89)

In article <776@east.East.Sun.COM> geoff@hinode.East.Sun.COM (Geoff Arnold @ Sun BOS - R.H. coast near the top) writes:
>In article <899@orbit.UUCP> jeb@pnet51.cts.com (James Borchart) writes:
>>        3: SUN's PC-NFS, running on a PC connected by ethernet to the
>>           ISC server, gets "authentication failure" no matter what is typed
>>           for the username and password, or if the user needs a password.
>I don't have a copy of the ISC stuff running here (lame excuse: I should
>order one), ...

I think that when we encountered this a while back we determined that pcnfsd
was not aware that the password field had been moved to /etc/shadow...








-- 
Charlie Sauer  Dell Computer Corp.     !'s:uunet!dell!sauer
               9505 Arboretum Blvd     @'s:sauer@dell.com
               Austin, TX 78759-7299   
               (512) 343-3310

geoff@hinode.East.Sun.COM (Geoff Arnold @ Sun BOS - R.H. coast near the top) (09/07/89)

In article <1989Sep6.175639.22619@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes:
>In article <1989Sep1.150312.1136@esegue.uucp> johnl@esegue.UUCP (John Levine) writes:
>>The problem is that the standard pcnfs daemon doesn't know that passwords
>>are in /etc/shadow...
>>I do have to wonder whether they ever bothered to test anything; it's a very
                              ^^^^ = ISC
>>obvious bug.
>
>It's well known that Sun's quality-control department consists of a single
>clerk with an IQ of 37, equipped with a rubber stamp that says "PASSED".

Henry, you weren't reading along this thread. We distribute pcnfsd
in source format. Since Sun doesn't use /etc/shadow, why should we
test it? ISC _does_ use /etc/shadow, and omitted to make the necessary
mods to pcnfsd. I'm sure they'll fix it - in fact the next distribution
of pcnfsd source will include this code #ifdef'd, courtesy the CTIX
developers. Don't blame our PC-NFS SQA for this, though. 

There's always something of a lag between releases of stuff - for example,
we haven't yet got a pcnsfd out there which supports the passwd.adjunct
security features in SunOS - but (1) we're working on it, and (2) the
source is there for anyone to customize or fix.

Geoff

Geoff Arnold,                              Internet: geoff@East.Sun.COM
PCDS Group, Sun Microsystems Inc.
---------------------------------------------------------------------------
Most bizarre title of the week: "LAN Standards - Do you Need Them?" (in BYTE)

henry@utzoo.uucp (Henry Spencer) (09/07/89)

In article <798@east.East.Sun.COM> geoff@East.Sun.COM (Geoff Arnold @ Sun BOS - R.H. coast near the top) writes:
>>It's well known that Sun's quality-control department consists of a single
>>clerk with an IQ of 37, equipped with a rubber stamp that says "PASSED".
>
>Henry, you weren't reading along this thread...

My apologies to Sun for blaming them for something that wasn't their fault.
That IQ-37 clerk has many things to answer for, but this wasn't one of them.
-- 
V7 /bin/mail source: 554 lines.|     Henry Spencer at U of Toronto Zoology
1989 X.400 specs: 2200+ pages. | uunet!attcan!utzoo!henry henry@zoo.toronto.edu