I get exactly the same behaviour on two of my Samba servers using winbind from
2.2.2 on Slackware 7.1 and 8.0 in a native mode win2k domain.
The last messages I see in the winbind log after wbinfo -u are:
[2001/10/24 11:37:29, 5] rpc_client/cli_pipe.c:rpc_api_pipe(408)
rpc_api_pipe: len left: 0 smbtrans read: 52
[2001/10/24 11:37:29, 6] rpc_client/cli_pipe.c:rpc_api_pipe(446)
rpc_api_pipe: fragment first and last both set
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_debug(63)
000018 samr_io_r_query_dispinfo
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint32(547)
0018 total_size : 0001b920
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint32(547)
001c data_size : 00000000
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint16(518)
0020 switch_level: 0001
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint32(547)
0024 num_entries : 00000000
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint32(547)
0028 ptr_entries : 0228a008
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint32(547)
002c num_entries2: 00000000
[2001/10/24 11:37:29, 6] rpc_parse/parse_prs.c:prs_debug(63)
000030 sam_io_sam_dispinfo_1 users
[2001/10/24 11:37:29, 5] rpc_parse/parse_prs.c:prs_uint32(547)
0030 status: 00000000
[2001/10/24 11:37:29, 5] nsswitch/winbindd.c:client_read(406)
read failed on sock 12, pid 14812: EOF
Mike
On Mon, Oct 22, 2001 at 04:44:33PM -0400, Sean O'Grady
wrote:> Greetings,
>
> I've done a 2.2.2 install from source on a Debian Woody release and
> have hit a wall (configure --with-winbind --with-pam). I've followed
> the Winbind setup doc and everything seems to have gone ok except I am
> not able to get Domain User or Group information from the DC.
>
> I did the smb.conf setup for winbindd with --
>
> ### WINBIND STUFF
> winbind separator = +
> winbind uid = 1001-20000
> winbind gid = 1001-20000
> winbind enum users = yes
> winbind enum groups = yes
> template shell = /bin/tcsh
> winbind cache time = 10
>
> security = domain
> password server = myhost
> workgroup = MYDOMAIN
>
> I am able to join the computer to the domain successfully and a wbinfo
> -t reports that the secret is good. However when I do a wbinfo -u or a
> wbinfo -g I recieve "Error looking up domain users" or
"Error looking up
> domain groups".
>
> The pam.d files have been modified to include --
>
> auth sufficient /lib/security/pam_winbind.so
> account required /lib/security/pam_winbind.so
>
> Which I believe is all that was required to be added for PAM. I've also
> made sure that /lib/libnss_winbind.so and the symlink are there and
> modified /etc/nsswitch.conf accordingly.
>
> I've run winbindd in debug mode attached to the terminal to see if
> something *pops up* at me and I did notice some moody behaviour being
> reported.
>
> I run winbindd -d 10 -i
>
> and send
>
> wbinfo -t
>
> I see this on the terminal
>
> contacting controller MYSERVER to check secret --
>
> secret is good
> read failed on sock 10, pid 5623: EOF
>
> Also something similiar happens with the wbinfo -u call --
>
> accepted socket 10
> [ 5622]: list users
> read failed on sock 10, pid 5622: EOF
>
> and with the wbinfo -g call --
>
> accepted socket 10
> [ 5626]: list groups
> read failed on sock 10, pid 5626: EOF
>
>
> The error "read failed" appears to be telling me something, but I
don't
> believe that I'm getting it ;)
>
>
> The extremely weird part of the matter is that the call wbinfo -n
> username actually returns a valid SID from the DC and sending the SID
> back returns the DOMAIN+username value for that SID, although the
"read
> failed" message still appears in the debug messafes while doing this.
>
> Any thoughts, questions, suggestions or help offered is appreciated.
>
> Thanks,
> Sean
>
> --
> Sean O'Grady
> Information Technology - SS
> Sheridan College
> 905-845-9430 x. 2166
> sean.ogrady@sheridanc.on.ca
>
> --
> To unsubscribe from this list go to the following URL and read the
> instructions: http://lists.samba.org/mailman/listinfo/samba
>