[Samba] Winbind(?) woes after hard disk full

Andi samba at faerun.fsnet.co.uk
Tue Oct 17 14:07:22 GMT 2006


Hi all.

Yesterday morning, we had a problem with our Samba box (3.0.20b-SUSE), 
traced to the root(/) partition becoming full. Upon restart, it was 
later reported that users at another office (trusted domain), could not 
access our server.

Everything had been running smoothly for over 5 months.

Running 'wbinfo -u/-g' returns a list of users/groups from the local 
domain, prefixed with the remote/trusted domain. No trusted-domain users 
or groups are listed.

Checking the trusts with 'wbinfo -t' succeeds.

It would appear that smbd and Winbind are having some problems (see log 
snippet below).

I'm considering rolling back the idmaps to an ldif copy I have from 
about 3 weeks ago, but am concerned that it is something else causing this.

Does anyone have any ideas?

Thx

Andi.


---- Syslog snippet
Oct 17 14:28:37 server smbd[11547]: [2006/10/17 14:28:37, 0] 
rpc_server/srv/netlog_nt.c:getmd4pw(242)

Oct 17 14:28:37 server smbd[11547]:   get_md4pw: Workstation MYDOMAIN$: 
No account in domain

Oct 17 14:29:37 server winbindd[2246]: [2006/10/17 14:29:37, 0] 
rpc_client/cli_pipe.c:rpc_api_pipe(438)

Oct 17 14:29:37 server winbindd[2246]:   cli_pipe: return critical 
error. Error was Call timed out: server did not respond after 10000 
milliseconds

Oct 17 14:29:37 server winbindd[2246]: [2006/10/17 14:29:37, 0] 
rpc_client/cli_pipe.c:cli_rpc_close(1767)

Oct 17 14:29:37 server winbindd[2246]:   cli_rpc_open failed on pipe 
\lsarpc to machine 172.16.1.7.  Error was Call timed out: server did no
t respond after 10000 milliseconds
---








More information about the samba mailing list