> Steven N. Hirsch wrote:
> > That's the one I'm using, and it is _not_ working. Wonder what the
> > difference is between my setup and the one that works?
>
> I'm not sure, but perhaps a few tests would help show what's happening.
> If you could try out these tests on a WfW share:
>
> touch foo
> chmod -w foo
> cp -a foo bar
> chmod +w foo
> echo hi >>foo
>
> and verify whether the permissions and timestamps are behaving, and what
> if any error messages occur.
Bill,
Very interesting. Check this out:
Script started on Sat Dec 6 00:50:30 1997
(hirsch@cy) ~ : cd /tmp/smb/TEMP/BENCH
(hirsch@cy) /tmp/smb/TEMP/BENCH : ls -al
total 31
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 .
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 ..
-rwxr-xr-x 1 hirsch system 23925 Nov 26 08:57 COPS.DOC
-rwxr-xr-x 1 hirsch system 4151 Oct 10 20:43 README
(hirsch@cy) /tmp/smb/TEMP/BENCH : touch foo
(hirsch@cy) /tmp/smb/TEMP/BENCH : ls -al
total 31
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 .
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 ..
-rwxr-xr-x 1 hirsch system 23925 Nov 26 08:57 COPS.DOC
-rwxr-xr-x 1 hirsch system 0 Dec 6 00:40 FOO
-rwxr-xr-x 1 hirsch system 4151 Oct 10 20:43 README
(hirsch@cy) /tmp/smb/TEMP/BENCH : chmod -w foo
(hirsch@cy) /tmp/smb/TEMP/BENCH : ls -al
total 31
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 .
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 ..
-rwxr-xr-x 1 hirsch system 23925 Nov 26 08:57 COPS.DOC
-r-xr-xr-x 1 hirsch system 0 Dec 6 00:40 FOO
-rwxr-xr-x 1 hirsch system 4151 Oct 10 20:43 README
(hirsch@cy) /tmp/smb/TEMP/BENCH : cp -a foo bar
(hirsch@cy) /tmp/smb/TEMP/BENCH : ls -al
total 31
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 .
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 ..
-r-xr-xr-x 1 hirsch system 0 Dec 6 00:41 BAR
-rwxr-xr-x 1 hirsch system 23925 Nov 26 08:57 COPS.DOC
-r-xr-xr-x 1 hirsch system 0 Dec 6 00:40 FOO
-rwxr-xr-x 1 hirsch system 4151 Oct 10 20:43 README
(hirsch@cy) /tmp/smb/TEMP/BENCH : chmod +w foo
(hirsch@cy) /tmp/smb/TEMP/BENCH : ls -al
total 31
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 .
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 ..
-r-xr-xr-x 1 hirsch system 0 Dec 6 00:41 BAR
-rwxr-xr-x 1 hirsch system 23925 Nov 26 08:57 COPS.DOC
-rwxr-xr-x 1 hirsch system 0 Dec 6 00:40 FOO
-rwxr-xr-x 1 hirsch system 4151 Oct 10 20:43 README
(hirsch@cy) /tmp/smb/TEMP/BENCH : echo hi >> foo
(hirsch@cy) /tmp/smb/TEMP/BENCH : ls -al
total 32
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 .
drwxr-xr-x 1 hirsch system 512 Nov 9 14:37 ..
-r-xr-xr-x 1 hirsch system 0 Dec 6 00:41 BAR
-rwxr-xr-x 1 hirsch system 23925 Nov 26 08:57 COPS.DOC
-rwxr-xr-x 1 hirsch system 3 Dec 6 00:51 FOO
^^^^^^^
-rwxr-xr-x 1 hirsch system 4151 Oct 10 20:43 README
(hirsch@cy) /tmp/smb/TEMP/BENCH : exit
Script done on Sat Dec 6 00:51:58 1997
Interesting how the mtime of "FOO" jumps by +11-minutes from cat'ing three
bytes to it, eh?
Here's the system logs:
Dec 6 00:49:58 cy kernel: smb_request_ok: rcls=1, err=18 mapped to 0
Dec 6 00:50:48 cy kernel: smb_request_ok: rcls=1, err=18 mapped to 0
Dec 6 00:51:15 cy kernel: smb_errno: class ERRDOS, code 50 from command 9
Dec 6 00:51:15 cy kernel: smb_errno: class ERRDOS, code 50 from command 9
Dec 6 00:51:15 cy kernel: smb_proc_settime: BENCH/bar failed, res=-5, rcls=1, err=50
Dec 6 00:51:17 cy kernel: smb_request_ok: rcls=1, err=18 mapped to 0
I hope this is of help. Do you suppose that having the system clock on
the server out of sync with the client (client lags by ~11 minutes) is
triggering this? If so, how can I keep them in sync? And, why would this
not cause similar problems against 2.0.32?
Steve