> More details:
> - Running in workgroup mode
> ...
> Oddities;
> idmap list/dump shows nothing. Is that normal? Think that used to show a 
> bunch of mappings.

idmap is not used in workgroup mode, seeing nothing is correct.

> If I try to browse the server from my vista pc, it looks like the smbd 
> process dies and restarts; it gets a new pid, and syslog shows:
> Nov  3 19:22:19 reaver smbsrv: [ID 852980 kern.notice] NOTICE: SmbOplocks: 
> disabled
> Any clues in that?

You can ignore the oplock message, it's just an informational thing.

If smbd is crashing, it should leave core files behind - per coreadm 
settings
or in the root directory.  The smbd core files would be useful.  If you find
them, run mdb on a core file and use $c to dump the stack, and post it here.

The original panic may have left core files in /var/crash/<hostname>
The stack from that panic may be useful.

Alan

----- Original Message ----- 
From: "Martin Gisch" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Monday, November 03, 2008 12:26 AM
Subject: [cifs-discuss] SMB shares broken after reboot


> Solaris Cifs Gurus,
>
> I've got a 2008.05 (snv_86) system. It's got a ZFS pool with a handful of 
> file systems on it that are shared over smb (Using the zfs sharesmb 
> options).
>
> It's been running perfectly for months.
>
> This weekend the server panic'd; something about memory corruption for 
> connection...
> It came back up ok, and my mapped drives seemed to recover.
> Today I manually rebooted it to run a memory test CD, no problems there.
> Bring the OS back up, and now I can't access any of the SMB shares 
> anymore. All attempts result in "SmbLogon[LAN\MARTIN]: WRONG_PASSWORD" 
> style errors.
>
> I havn't changed any passwords anywhere, either on client or server. The 
> password files under /var/smb had modification stamps months ago.
>
> I have no idea whats gone wrong or how to gather more debug info from the 
> cifs processes to figure it out. Can anyone suggest a course of action?
>
> More details:
> - Running in workgroup mode
> - Main client is a Vista PC
>
> [EMAIL PROTECTED]:/var/smb# sharemgr show -vp
> default nfs=()
> smb smb=()
> zfs
>    zfs/rpool/export smb=()
>          export=/export
>    zfs/store/movies smb=()
>          Movies=/store/movies
>    zfs/store/music smb=()
>          Music=/store/music
>    zfs/store/overlord nfs=() nfs:sys=(rw="overlord.home.mg77.net")
>          /store/overlord
>    zfs/store/tv smb=()
>          TV=/store/tv
>
> (NFS seems to be working ok)
>
> [EMAIL PROTECTED]:/var/smb# zfs get sharesmb store/tv
> NAME      PROPERTY  VALUE     SOURCE
> store/tv  sharesmb  name=TV   local
>
> [EMAIL PROTECTED]:/var/smb# ps -ef | grep smb
>    root  1385     1   0 19:07:14 ?           0:00 /usr/lib/smbsrv/smbd 
> start
>    root  1429   540   0 19:19:41 pts/1       0:00 grep smb
>
> [EMAIL PROTECTED]:/var/smb# idmap list
> [EMAIL PROTECTED]:/var/smb# idmap dump
> [EMAIL PROTECTED]:/var/smb#
>
> Oddities;
> idmap list/dump shows nothing. Is that normal? Think that used to show a 
> bunch of mappings.
> If I try to browse the server from my vista pc, it looks like the smbd 
> process dies and restarts; it gets a new pid, and syslog shows:
> Nov  3 19:22:19 reaver smbsrv: [ID 852980 kern.notice] NOTICE: SmbOplocks: 
> disabled
> Any clues in that?
>
> Thanks for any advice.
> -Martin.
> -- 
> This message posted from opensolaris.org
> _______________________________________________
> cifs-discuss mailing list
> [email protected]
> http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
> 

_______________________________________________
cifs-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/cifs-discuss

Reply via email to