> >   5) Do you know what "locale" is? Do you know how to change it?
> >      (this help to avoid the russian error message) 
> 
> Well, I tried to type
> # LC_ALL="en_US" smbnetfs
> but this didn't help: errors were in russian
> Now I've dpkg-reconfigured locales packages and genetated english locale
> files.
> Well, be sure I can find out english messages.
> I won't post russian messages any more, excuse me.

It's better use LANG=C, this locale is always present.

 
> > Then, I need to know your samba, libsmbclient and smbnetfs configuration
> >   1) samba configuration (contents of the file /etc/samba/smb.conf). It's 
> >      better to provide the output of "testparm" utility, because smb.conf
> >      contain to much trash
> 
> OK, here testparm output follows:
> [global]
>       dos charset = cp866
>       unix charset = KOI8
>       workgroup = HACKERS
>       server string = %h server
>       security = SHARE
>       obey pam restrictions = Yes
>       passdb backend = tdbsam
>       passwd program = /usr/bin/passwd %u
>       passwd chat = *Enter\snew\sUNIX\spassword:* %n\n
> *Retype\snew\sUNIX\spassword:* %n\n *password\supdated\ssuccessfully* .
> syslog = 0 log file = /var/log/samba/log.%m
>       max log size = 1000
>       os level = 50
>       preferred master = No
>       local master = No
>       domain master = No
>       dns proxy = No
>       wins server = 10.0.0.1
>       panic action = /usr/share/samba/panic-action %d
>       invalid users = root

That's Ok, I really need to get the following from this:
 -- domain configuration (you are not in domain)
 -- master browser settings (you are not a master browser)
 -- binding of interface (all interface)
 -- netbios name resolv order (lmhosts host wins bcast)

> >   2) libsmbclient configuration (contents of the file ~/.smb/smb.conf
> >   from the
> >      user what run smbnetfs)
> 
> This file does not exist

That's bad, especially for russia. As result libsmbclient works in koi8-r.
It's not suitable to feet all unicode characters to koi8-r. You are waiting
for a problems.

Put to your ~/.smb/smb.conf something similar to (this is usually sufficient
for samba network without domain) 

[global]
    dos charset = cp866
    display charset = utf-8
    workgroup = HACKERS
    map to guest = Bad User
    deadtime = 5
#   socket options = TCP_NODELAY IPTOS_LOWDELAY SO_RCVBUF=8192 SO_SNDBUF=8192
    create mask = 0640
    directory mask = 0750
    case sensitive = No
    wins server = 10.0.0.1

after that you should set samba_charset=utf-8 in your ~/.smb/smbnetfs.conf
 
> >   3) smbnetfs configuration (contents of the file ~/.smb/smbnetfs.conf
> >   from the
> >      user what run smbnetfs)
> 
> I've posted it in the very first message

I miss it during the first reading. It's quite ok, just setup libsmbclient to
utf-8 as i wrote above.

> > Can you run smbnetfs in debug mode and repeat your test? 
> > There are 2 possibility
> >   1) turn on smbnetfs log (set smbnetfs_debug level). Put the
> >      following line to your ~/.smb/smbnetfs.conf 
> > 
> >      smbnetfs_debug          6
> > 
> >      Possible values are from 0 to 10, but the value of 6 is
> >      usually enough.
> > 
> >   2) Turn on samba log (set smb_debug level). Put the following line
> >      to your ~/.smb/smbnetfs.conf
> > 
> >      smb_debug          number
> > 
> >      where the number is a value from 0 to 10 (no any recommendation,
> >      in hard situation i usually use 10)

It make sense to use smb_debug=10 when you locate the place of error quite
close. If there are no errors -- it's not help.


> > Sometime the reason of such behavior is the broken tdb database of your
> > samba server. In this case it helps to delete all "*.tdb" files from
> > "/var/cache/samba" and it's subdirectories, then restart samba and
> > winbind (this is safe operation, this files will be recreated)
> 
> Could you tell me, how samba _server_ could cause _client_ crashes?

I know an only scenario. libsmbclient search some information in samba
server database (it created and managed by samba server). Sometime this
database gets damaged by unknown reason. So when you start working with
it you may face with null pointer call, infinite loops and so on.
This will happen only on certain queries, so you will face a problem
any time you connect to host that relate to broken record. It's most 
dangerous when the master browser record is damaged. In this case you
will see an error on every network scanning

Mikhail Kshevetskiy

PS: it's possible to turn off network scanning through ~/.smb/smbnetfs.conf
  



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to