[comp.protocols.nfs] Sun PC-NFS 3.5 "ls -l" output changes

ado@elsie.nci.nih.gov (Arthur David Olson) (05/14/91)

+ From ado Mon May 13 13:09:57 1991
+ To: sunbugs@sun.com
+ Subject: Sun PC-NFS 3.5 "ls -l" output changes
+ 
+ Description:
+ 	The Sun PC-NFS 3.5 "ls -l" output changes depending on whether
+ 	you have or have not done a "net name".
+ 
+ Repeat-By:
+ 	Note the "nobody"s that appear on the lines highlighted with
+ 	"A>>>>" below as distinct from the "-2"s that appear on the lines
+ 	highlighted with "B>>>>" below.
+ 	
+ 	C:\>rdate elsie
+ 	Mon May 13 16:56:10 1991
+ 	C:\>net -v
+ 	Current version is 3.5.
+ 	C:\>net use k: \\elsie\home\pcs\petunia\programs
+ 	C:\>net name ado *
+ 	C:\>mkdir k:playpen
+ 	C:\>k:
+ 	K:\>cd playpen
+ 	K:\PLAYPEN>echo hi > alpha
+ 	K:\PLAYPEN>echo hi > beta
+ 	K:\PLAYPEN>net logout
+ 	K:\PLAYPEN>echo hi > gamma
+ 	K:\PLAYPEN>echo hi > delta
+ 	K:\PLAYPEN>ls -l
+ 	drwxrwxrwx  2 ado             - May 13 12:56 .
+ 	drwxr-xr-x 14 ado             - May 13 12:56 ..
+ 	-rw-rw-rw-  1 ado             5 May 13 12:56 alpha
+ 	-rw-rw-rw-  1 ado             5 May 13 12:56 beta
+ A>>>>	-rw-rw-rw-  1 nobody          5 May 13 12:56 gamma
+ A>>>>	-rw-rw-rw-  1 nobody          5 May 13 12:56 delta
+ 	K:\PLAYPEN>net name ado *
+ 	K:\PLAYPEN>ls -l
+ 	drwxrwxrwx  2 ado             - May 13 12:56 .
+ 	drwxr-xr-x 14 ado             - May 13 12:56 ..
+ 	-rw-rw-rw-  1 ado             5 May 13 12:56 alpha
+ 	-rw-rw-rw-  1 ado             5 May 13 12:56 beta
+ B>>>>	-rw-rw-rw-  1 -2              5 May 13 12:56 gamma
+ B>>>>	-rw-rw-rw-  1 -2              5 May 13 12:56 delta
+ -- 
+ 		Arthur David Olson	ado@elsie.nci.nih.gov
+ 		ADO and Elsie are Ampex and Borden trademarks

geoff@bodleian.East.Sun.COM (Geoff Arnold @ Sun BOS - R.H. coast near the top) (05/14/91)

Quoth ado@elsie.nci.nih.gov (Arthur David Olson) (in <1478@nih-csl.nih.gov>):
#+ From ado Mon May 13 13:09:57 1991
#+ To: sunbugs@sun.com
#+ Subject: Sun PC-NFS 3.5 "ls -l" output changes
#+ 
#+ Description:
#+ 	The Sun PC-NFS 3.5 "ls -l" output changes depending on whether
#+ 	you have or have not done a "net name".

Yes, all of this is quite consistent :-) The algorithm we use
to interpret the uid is as follows: if the uid is the same
as the currently loged-in user, we use the username stored
at the time we did the NET NAME. Otherwise if NIS is running,
we do a lookup in passwd.byuid; if we get a hit, we cache the result
and use the corresponding username. Otherwise we use the numeric
uid. In Arthur's example, it looks like there is no "nobody" entry
in the local NIS map.

Geoff

-- Geoff Arnold, PC-NFS architect, Sun Microsystems. (geoff@East.Sun.COM)   --
------------------------------------------------------------------------------
--     Sun Microsystems PC Distributed Systems ...                          --
--            ... soon to be a part of SunTech (stay tuned for details)     --

ado@elsie.nci.nih.gov (Arthur David Olson) (05/14/91)

> . . .it looks like there is no "nobody" entry in the local NIS map.

Thanks to Geoff for the quick followup. . .and the accurate diagnosis.

The good news is that adding a "nobody" entry to the NIS map produces
consistent results.

The other news is that the entry can't be the
	nobody:*:65534:65534::/:
line of SunOS 4.1.1's /etc/passwd file; you get to use
	nobody:*:-2:-2::/:
to get consistent "ls -l" output on PCs.
-- 
		Arthur David Olson	ado@elsie.nci.nih.gov
		ADO and Elsie are Ampex and Borden trademarks