Suse 13.2/64, ntfs-3g partition, fstab mount owner gets "access is denied" from Samba on win7

This is a fresh install of Suse 13.2 with Samba from distro. The only hard drive has a / partition with btrfs and a large, existing ntfs partition. The mount is owned specifically by the user, david:id=1000
dojpc:/ # grep ntfs /etc/fstab
UUID=D6FE6724FE66FBDD /ntfs ntfs-3g users,uid=1000,gid=users,fmask=002,dmask=002 0 0
david@dojpc:~/pic> id
uid=1000(david) gid=100(users) groups=100(users)

There is a symlink in /home/dave/pic. Everything from the fstab mount to the root directory for the mount (/ntfs) to the pic directory is 100% owned by david

david@dojpc:~/pic> ll ~ | grep piclrwxrwxrwx 1 david users 9 Feb 10 12:09 pic -> /ntfs/pic
[FONT=Verdana]david@dojpc:~/pic> ll / | grep ntfs
drwxrwxr-x 1 david users 16384 Feb 7 18:28 ntfs
david@dojpc:~/pic> ll ~/pic/b* | head -n1
-rwxrwxr-x 1 david users 4377296 Aug 2 2012 /home/david/pic/bardot.BPB_0282.jpg

The user has full access in the /ntfs/pic directory:
david@dojpc:~/pic> ll xyz
ls: cannot access xyz: No such file or directory
david@dojpc:~/pic> touch xyz
david@dojpc:~/pic> ll xyz
-rwxrwxr-x 1 david users 0 Feb 10 14:26 xyz
david@dojpc:~/pic> rm xyz
david@dojpc:~/pic> ll xyz
ls: cannot access xyz: No such file or directory

The Samba which came on the install dvd (Samba version 4.1.12-2.1-3307-SUSE-oS13.2-x86_64) is set to “follow symlinks = yes” and use “wide links = yes”


[global]
    passdb backend = tdbsam
    cups options = raw
    netbios name = dojpc
    include = /etc/samba/dhcp.conf
    usershare allow guests = No
    printcap cache time = 750
    os level = 20
    printing = cups
    default = homes
    valid users = brianp,david
    usershare max shares = 100
    map to guest = Bad User
    logon drive = P:
    encrypt passwords = yes
    logon path = \\%L\profiles\.msprofile
    logon home = \\%L\%U\.9xprofile
    workgroup = FRACTASIA
    printcap name = cups
    security = user
    default service = homes
    wins support = Yes
***        follow symlinks = yes                              <<  both directives deleted by testparm***
***        wide links = yes***
[homes]
    inherit acls = Yes
    comment = Home Directories
    valid users = %S, %D%w%S
    writable = yes
[profiles]
    comment = Network Profiles Service
    path = %H
    read only = No
    store dos attributes = Yes
    create mask = 0600
    directory mask = 0700
[users]
    veto files = /aquota.user/groups/shares/
    inherit acls = Yes
    path = /home
    comment = All users
    valid users = david,brianp
    writeable = yes
[groups]
    comment = All groups
    path = /home/groups
    read only = No
    inherit acls = Yes
[printers]
    comment = All Printers
    path = /var/tmp
    printable = Yes
    create mask = 0600
    browseable = No
[print$]
    comment = Printer Drivers
    path = /var/lib/samba/drivers
    write list = @ntadmin root
    force group = ntadmin
    create mask = 0664
    directory mask = 0775


[pic]
    path = /ntfs/pic
[/FONT][FONT=Verdana]writable = yes

[/FONT]

[FONT=Verdana]

[/FONT]==================================
The testparm program appears to mangle, reformat and outright delete essential info. Comparing the testparm dump to the actual /etc/samba/smb.conf, here is a list of the singular lines appearing in only 1 of the 2 sources:

C:/bin>freq d:/tmp/samba.txt -cts # Frequency analysis, -Compress all whitespace, -Trim front/end space, -Singular (only seen once)::
Can’t find include file /etc/samba/dhcp.conf << file missing entirely. Using host list with static IPs.
Load smb config files from /etc/samba/smb.conf
Server role: ROLE_STANDALONE
default = homes
encrypt passwords = yes
follow symlinks = yes <<< In the smb.conf only, forgotten/mangled/foobarred by testparm
idmap config * : backend = tdb
include = /etc/samba/dhcp.conf
netbios name = dojpc << ping dojpc works on windoz
os level = 20
passdb backend = tdbsam
security = user
usershare allow guests = No
wide links = yes << destroyed by testparm
writable = yes
write list = @ntadmin, root

When I setup a totally separate share of /ntfs/pic mapping the network drive as \DOJPC\david with the suse password for user david. It supports both read and write and works well on Suse and windows \

[FONT=Verdana]d;/pic\friends>copy nb-2013.0709-158878.jpg z: => 1 file(s) copied.
d:/pic\friends>ll z:/nb-2013.0709-158878.jpg
-rw-rw-rw- 1 user group 2454457 Feb 8 2014 z:/nb-2013.0709-158878.jpg
d:/pic\friends>rm z:/nb-2013.0709-158878.jpg
d:/pic\friends>ls z:/nb-2013.0709-158878.jpg
ls: z:/nb-2013.0709-158878.jpg: No such file or directory

It appears that somehow the “follow symlinks = yes” and “wide links = yes” directives are being ignored. [/FONT]
[FONT=Verdana]
Is it possible to traverse a symbolic link in Linux to an NTFS partition via Samba as seen by windows?

[/FONT]

Temporarily disable apparmour to see if it’s an issue:


systemctl stop apparmor

And try to access the files.

There is no difference before or after systemctl stop apparmor run as root:
Y:/pic is not accessible
Access is denied

There was no output so I believe that means that there was nothing to report.

A weird thing is that the YAST service manager lists the apparmor service as being both Disabled and Active.

I am very sure I did not monkey with it before running this test to disable it. How did it get ACTIVE if it has been Disabled?

C:/bin>net use < snip >
OK Y: \dojpc\david Microsoft Windows Network
OK Z: \DOJPC\pic Microsoft Windows Network

C:/bin>touch z:/apparmor.text.txt

david@dojpc:~/pic> uptime
19:28pm up 6:08, 9 users, load average: 0.02, 0.03, 0.05
david@dojpc:~/pic> uptime >> /ntfs/pic/apparmor.text.txt
19:29pm: command not found

C[THIS_IS_NOT_A_FROWNIE_FACE]bin>cat z:/apparmor.text.txt
uptime
19:28pm up 6:08, 9 users, load average: 0.03, 0.03, 0.05

C:\bin>echo Hi from win >> z:/apparmor.text.txt
david@dojpc:~/pic> cat /ntfs/pic/apparmor.text.txt
uptime
19:28pm up 6:08, 9 users, load average: 0.03, 0.03, 0.05
Hi from win

david@dojpc:~/pic> ll /ntfs | grep pic
drwxrwxr-x 1 david users 12288 Feb 10 19:25 pic

david@dojpc:~/pic> ll /home/david/ | grep pic
lrwxrwxrwx 1 david users 9 Feb 10 12:09 pic -> /ntfs/pic

I can get to /ntfs/pic through the direct Samba share, but not through the symbolic link ~david/pic -> /ntfs/pic

brianpaustin wrote:

>
> There is no difference before or after systemctl stop apparmor run as
> root:
> Y:/pic is not accessible
> Access is denied
>
> There was no output so I believe that means that there was nothing to
> report.
<snip>
>
> I can get to /ntfs/pic through the direct Samba share, but not through
> the symbolic link ~david/pic -> /ntfs/pic
>

brianpaustin;

You need to set the parameter “allow insecure wide links” to true in the
global section of /etc/samba/conf.


allow insecure wide lengths = yes

See the write up on this parameter man smb.conf. Also the parameter “follow
symlinks” and “wide links” are share level parameters not global parameters.


P.V.
“We’re all in this together, I’m pulling for you” Red Green

>> You need to set the parameter “allow insecure wide links” to true in the
global section of /etc/samba/conf.

<< samba.org/samba/docs/man/manpages-3/smb.conf.5.html
>>wide links which allows the server to follow symlinks outside of a share path is automatically disabled

There were 2 share paths pointing to the pic directory, a direct path via [pic] -> path = /ntfs/pic and a symbolically linked path via [homes]/<user=david>/pic -> /ntfs/pic. The link which was causing problems was under a share being exported.

This partially explains why the smb.conf vs testparm diff showed “follow symlinks” and “wide links” missing from the testparm spew.

When TESTing a servers PARaMeters, only giving you the good news does not seem like the full story. I would have expected some error message when config errors were detected rather than silent ignoration:
HEY BONEHEAD: Your Global PARMs “follow symlinks” and “wide links” need to be moved to a share or they will be ignored!

I updated the smb.conf and used yast to stop/start smb/nmb:
dojpc:/etc/samba # diff smb.conf smb.2015.0211.conf
26c26,27
< allow insecure wide links = Yes

> follow symlinks = yes
> wide links = yes
66,67d66
< follow symlinks = yes
< wide links = yes

More bad news:
==> /var/log/samba/log.smbd <==
[2015/02/11 09:00:52.488931, 0] …/source3/param/loadparm.c:5497(widelinks_warning)
Share ‘david’ has wide links and unix extensions enabled. These parameters are incompatible. Wide links will be disabled for this share.

So, the “Wide Link” pointing from the [home] share to a foreign partition not on this share can not possibly work with wide links disabled. Checking …

X;/pic>net use
OK X: \dojpc\david Microsoft Windows Network
OK Z: \DOJPC\pic Microsoft Windows Network

C;/bin>x:
X;/pic>cp z:/BPB_1835.jpg ./BPB_1835.test.jpg

X;/pic>ll z:/BPB_1835.jpg ./BPB_1835.test.jpg
-rw-rw-rw- 1 user group 2553637 Feb 11 09:54 ./BPB_1835.test.jpg
-rw-rw-rw- 1 user group 2553637 Mar 15 2003 z:/BPB_1835.jpg

X;/pic>rm ./BPB_1835.test.jpg
X;/pic>rm ./BPB_1835.test.jpg
rm: ./BPB_1835.test.jpg: No such file or directory

The log files clearly state that wide links are disabled. Testing proves this is not the current behavior. Who is misinformed?

It appears that the boot-time Samba logs go to /var/log/samba. Yast → servers → stop/start smb/nmb does not show up here. tail -f /var/log/samba/log.smbd does not budge when yast → services restarts smb??? Inquiring minds want to know…

When config errors are found they are silently ignored.
Serious looking error messages in the server logs (“Wide links will be disabled”) can be ignored???

And Webmin does not work. It will neither stop nor start the servers and the log file dir is /var/log/samba, not yast’s undisclosed location. The log file location is not specified either on the command line nor in the config file and the output is apparently being squirreled away by YAST.

It would be most convenient to look at the ps and know where the logs are rather than looking at /etc/menus/yast-settings.menu/initscripts/packages/default/autoyast/config/control/servers/optional/setup/suse/site/preferences/redirect/vendor/samba.cfg for the rest of the story :slight_smile:

Winders can now thrash all of David’s files. It’s working!!!

/var/log/messages

2015-02-11T20:24:44.061731+02:00 hircine systemd[1]: Stopping Samba SMB Daemon...

Depends on whether you consider wide links as a “serious error”, I would not as wide links is not really good security practice in general.

Webmin does not ship with the distribution and thus is not really supported.

But wait, there is NO /var/log/messages
dojpc:~ # cat /var/log/messages
cat: /var/log/messages: No such file or directory
YAST is getting overly creative!

Wherever YAST is stashing it, its well hidden:
dojpc:~ # locate samba | grep /log/
/var/log/samba
/var/log/samba/cores
/var/log/samba/log.nmbd
/var/log/samba/log.smbd

https://forums.opensuse.org/showthread.php/428805-Samba-Log-Files

… By default, smbd creates the log files in /var/log/samba/log.smbd …
P. V. “We’re all in this together, I’m pulling for you.” Red Green

This P. V. character sounds authoritative!

https://en.opensuse.org/openSUSE:Bugreport_Samba

Samba bug reporting and advanced debugging information
Providing files as attachment
Provide all log files from /var/log/samba/ directory …

Everybody seems to be in agreement where Samba log files belong, except for yast

Sending part of the log to the “industry standard” location and hiding the rest where not even MiukuMac can find it does not seem like the best accounting method, but it is the easiest for YAST. Having to first find and then merge various, randomly scattered log segments to get the full server story seems convoluted to put it charitably.

How about a command line argument which overrides any config file parm, makes it abundantly clear where the current output is going, and prevents yast from logjacking your data?!

Depends on whether you consider wide links as a “serious error”

A possible definition of a “serious error” would be when you order a specific, essential function and are completely ignored. Doing the best you can with imperfect information is good and robust. Detecting an error and making your commander check up on you progress in following orders rather than reporting the error is less than responsible.

wide links is not really good security practice in general.

Do you lock your screen at home when you get up to fetch a glass of milk? Most probably don’t. If you leave your office without locking your screen at IBM, you are summarily debadged and escorted to the nearest exit. Security is relative.

For a publically available server, I would concur with your security analysis. For this specific case, there is only one hard drive, one user, one sharing, home email / browser station, behind a firewall and the “wide” share is on a defunct windoz data partition. It has Facebook type pictures on it; nothing worth stealing. Size restrictions mandate use of the large partition rather than wasting 98.6% of a 1 TB drive.

Telling an unsophisticated user that all of your stuff is right here in your david directory except you pictures. For security reasons, you must open up a separate explorer window, navigate to the root directory, click on ntfs and then find PIC. “But I don’t need any security on my cell phone snaps. Can you fix it?” “Hey Brian! What happened to all of my pictures. I keep getting ACCESS DENIED when I click on PIC. Where did you hide them again???”

Webmin does not ship with the distribution and thus is not really supported.

I found Webmin v1.730-1 in YAST under software management on OpenSuse 13.2 ((Harlequin) (x86_64)
Perhaps YAST could spy on it to get pointers on where to stash log files!

The morale of the story is use Suse for what it is, a badass Linux OS. Run the servers you need from the server experts, the Apache/Samba/Mysql/… gangs, with their administration tools. Avoid the me too, add-on, dumbed down, least common denominator, maintenance focused; functionality maybe admin tools and “value additions” which scramble all of the standard stuff so that you are locked into using their third rate set of tools. Where did yast hide my data?

Customers want to make an informed decision:
Disclaimer: If all you need is the most basic functionality and ease of use is your top priority, the stripped-down Samba included with this distro may be just what you need. If you need a bit more control or features not found on our austere interface, a dedicated tool such as [SWAT->link::http://swat.net] or [WEBMIN-> link::www.[b]webmin.com] may work out better for you…

An OS company can’t be expected to be on the leading edge for every possible server. To paraphrase Eastwood,
A distro’s got to know its limitations.

On 2/11/2015 6:46 PM, brianpaustin wrote:
>
> Miuku;2694440 Wrote:
>>>
> Code:
> --------------------
> > > /var/log/messages
> > >>2015-02-11T20:24:44.061731+02:00 hircine systemd[1]: Stopping Samba SMB Daemon…
> --------------------
>>>
>
> But wait, there is NO /var/log/messages
> dojpc:~ # cat /var/log/messages
> cat: /var/log/messages: No such file or directory
> YAST is getting overly creative!
>
> Wherever YAST is stashing it, its well hidden:
> dojpc:~ # locate samba | grep /log/
> /var/log/samba
> /var/log/samba/cores
> /var/log/samba/log.nmbd
> /var/log/samba/log.smbd
>
> https://forums.opensuse.org/showthread.php/428805-Samba-Log-Files
>> … By default, smbd creates the log files in
>> -/var/log/samba-/log.smbd …
>> P. V. “We’re all in this together, I’m pulling for you.” Red Green
>>
> This P. V. character sounds authoritative!
>
> https://en.opensuse.org/openSUSE:Bugreport_Samba
>> Samba bug reporting and advanced debugging information
>> Providing files as attachment
>> Provide all log files from -/var/log/samba-/ directory …
>>
> Everybody seems to be in agreement where Samba log files belong, except
> for yast
>
> Sending part of the log to the “industry standard” location and hiding
> the rest where not even MiukuMac can find it does not seem like the best
> accounting method, but it is the easiest for YAST. Having to first find
> and then merge various, randomly scattered log segments to get the full
> server story seems convoluted to put it charitably.
>
> How about a command line argument which overrides any config file parm,
> makes it abundantly clear where the current output is going, and
> prevents yast from logjacking your data?!
>
>
>> Depends on whether you consider wide links as a “serious error”
>
> A possible definition of a “serious error” would be when you order a
> specific, essential function and are completely ignored. Doing the best
> you can with imperfect information is good and robust. Detecting an
> error and making your commander check up on you progress in following
> orders rather than reporting the error is less than responsible.
>
>
>> wide links is not really good security practice in general.
> Do you lock your screen at home when you get up to fetch a glass of
> milk? Most probably don’t. If you leave your office without locking your
> screen at IBM, you are summarily debadged and escorted to the nearest
> exit. Security is relative.
>
> For a publically available server, I would concur with your security
> analysis. For this specific case, there is only one hard drive, one
> user, one sharing, home email / browser station, behind a firewall and
> the “wide” share is on a defunct windoz data partition. It has Facebook
> type pictures on it; nothing worth stealing. Size restrictions mandate
> use of the large partition rather than wasting 98.6% of a 1 TB drive.
>
> Telling an unsophisticated user that all of your stuff is right here in
> your david directory except you pictures. For security reasons, you must
> open up a separate explorer window, navigate to the root directory,
> click on ntfs and then find PIC. “But I don’t need any security on my
> cell phone snaps. Can you fix it?” “Hey Brian! What happened to all of
> my pictures. I keep getting ACCESS DENIED when I click on PIC. Where did
> you hide them again???”
>
>
>> Webmin does not ship with the distribution and thus is not really
>> supported.
> I found Webmin v1.730-1 in YAST under software management on OpenSuse
> 13.2 ((Harlequin) (x86_64)
> Perhaps YAST could spy on it to get pointers on where to stash log
> files!
>
>
> The morale of the story is use Suse for what it is, a badass Linux OS.
> Run the servers you need from the server experts, the
> Apache/Samba/Mysql/… gangs, with their administration tools. Avoid the
> me too, add-on, dumbed down, least common denominator, maintenance
> focused; functionality maybe admin tools and “value additions” which
> scramble all of the standard stuff so that you are locked into using
> their third rate set of tools. Where did yast hide my data?
>
> Customers want to make an informed decision:
> Disclaimer: If all you need is the most basic functionality and ease of
> use is your top priority, the stripped-down Samba included with this
> distro may be just what you need. If you need a bit more control or
> features not found on our austere interface, a dedicated tool such as
> [SWAT->link::http://swat.net] or [WEBMIN-> link::www.*webmin.com] may
> work out better for you…
>
> *An OS company can’t be expected to be on the leading edge for every
> possible server. To paraphrase Eastwood,
> A distro’s got to know its limitations.
>
>
brianpaustin,

With 13.2 openSUSE does not install syslog-ng by default. The system logs are kept by the systemd-journald.service and
can be read with:


journalctl

system-ng is still available and can be installed to get back the earlier behavior.

If you are nervous about allowing insecure wide links you can turn off “unix extensions” in the [global] section of
/etc/samba/smb.conf.


unix extensions = no

This may break nix clients. The problem with wide links is that it is possible for a nix client to create a link to an
arbitrary location of the servers file system. See: man smb.conf

AFAIK YaST is not responsible for logging. It is merely a configuration tool. YaST does not put logs anywhere.


P.V.
“We’re all in this together, I’m pulling for you” Red Green

Ah yes, journalctl (or more aptly systemlis the first thing I disable nowadays due to its speed when it comes to handling large logs. In order to get it, you need to use rsyslog or similar.

What industry standard would that be? The fact that every distribution handles various log locations and configurations differently.

Take Debian for example, their Apache default webroot is in /var/www, while on SUSE it’s /srv/www and other distributions can even place it in /usr/local or /opt.

You realise the fact that it ignores it isn’t due to SUSE configuration? It’s in the Samba software itself - why don’t you install Ubuntu, Fedora or any other number of distributions and do the same thing there? I’ll guarantee you the service will ignore your misconfiguration and still work fine.

Webmin does not ship with the distribution, no matter what you say.

Feel free to look at the distribution source rpms or the DVD. Oh look at that, no Webmin. How curious.

You are not a customer because you didn’t pay for it. Once you do, you can call yourself a customer - now you are just a normal user who assumes too much, complains about things they get for free and contribute nothing back to and whine. Suggestion for the future; whine less, you’ll be received better in life. Sage advice.

The fact that you don’t know what you’re doing isn’t the distributions fault - sorry.

Here’s another quote for you; Only a fool blames his tools.

“Industry Standard” location for Samba files?

What industry standard would that be?

http://www.oreilly.com/openbook/samba/book/ch09_01.html
Using Samba by Robert Eckstein, David Collier-Brown, Peter Kelly
Here is an up-to-date configuration file that covers each of the share and logging options that we’ve seen so far:[global]
log file = /var/log/samba.log.%m*https://www.samba.org/samba/docs/using_samba/ch12.html
Chapter 12. Troubleshooting Samba
*You can specify a log directory to use with the -l flag on the command line when starting the Samba daemons. For example:# smbd -l /var/log/samba

**nmbd -l /var/log/samba

[though in the example their source compiled version goes in /usr/local/samba, they place the logs under ~samba/var]

Again quoting P. V. (didn’t you read this last time? :slight_smile:
https://forums.opensuse.org/showthread.php/428805-Samba-Log-Files

Mark;
It depends a bit on how much info you need. By default, smbd creates the log

files in /var/log/samba/log.smbd … P. V. …

**
*http://unix.stackexchange.com/questions/7681/how-can-i-log-samba-events

By default, logs are placed in samba_directory /var/smbd.log and samba_directory /var/nmbd.log

With a source install, logs are placed under [local_samba_root] /var (and with the tiny number of files involved the /log is unnecessary]

Ubuntu:
http://serverfault.com/questions/484442/samba-logging-all-to-get-logs-about-all-user-activities-on-shared-folders-and-fi

Samba logging all to get logs …

My /etc/samba/smb.conf contains

log level = 3
log file = /var/log/samba/log.%p

*http://serverfault.com/questions/389166/how-to-debug-samba-authorization-authentication-procedure
How to debug Samba, 1 Answer

watch log files - usually in /var/log/samba/

Here is some poor slob with his samba logs hijacked by some pin headed sysadmin who wants to watch tail one file all day and make believe he is in The Matrix:
https://www.redhat.com/archives/rhl-list/2004-July/msg05737.html
Samba in var/log/messages ???]

If i look in var/log/messages I see these lines …
Jul 24 20:43:11 Fedora nmbd[2279]: Unable to sync browse lists …
I guess that it is something connected with samba
And this is my samba.conf file

log file = /var/log/samba/%m.log

Can you relate to this schmuck?

Gen Too?
http://wiki.gentoo.org/wiki/Samba/HOWTO

The main Samba configuration file is /etc/samba/smb.conf …

log file = /var/log/samba/log.%m

*Your anarchistic insinuation that there are no standards and any file belongs anywhere makes me want to dump all of my meticulously sorted **file cabinets in the middle of my living room and run my leaf blower through them so I can [eventually] find everything in one place!

You realise the fact that it ignores it isn’t due to SUSE configuration?

I know this is not Suse. It just defeats the purpose of having a “Control System” when the more important half of your feedback loop is silenced; the ERRORS are what you want to see first before the trivial echoing of the part of the command that worked. This needs to be fixed!

Customers want to make an informed decision True or False? Duh!??

You are not a customer

http://www.thefreedictionary.com/customer
***2. ** Informal An individual with whom one must deal: That teacher is a tough customer.
Could it be that you are even less familiar with the English Language than you are with log files or the concept of STANDARDS?

Are we having a minor melt-down? Are you supposed to call childish names throw temper tantrums in writing when somebody wonders about the sanity of having having half a log file in the correct, standard location and the rest of it on some generic message dump? Is this really worth loosing your cool or do you have other issues which are hampering your ability to control your emotions.

I have my link working and my log files right where I want them. I’m cool.
I am left wondering what your contemptous characterization of users as “Pidgeons” signifies in contrast to your self-representation as a cute little bunny.

Only a fool blames his tools.

you are unprofessional

P. V.,

With 13.2 openSUSE does not install syslog-ng by default. The system logs are kept by the systemd-journald.service and
can be read with:

journalctl

system-ng is still available and can be installed to get back the earlier behavior.

Sending the Samba log at boot time to the “so called” “industry standard” /var/log/samba and then suddenly redirecting it to BFE when the yast->service_mgr restarts it can not by any stretch of the imagination be considered correct behavior. There are good arguments for A and arguments for B, but half A - half B is schizophrenic. This smells like a sev 1 bug.

If you are nervous about allowing insecure wide links you can turn off “unix extensions” …
This may break nix clients. The problem with wide links is that it is possible for a nix client to create a link to an
arbitrary location of the servers file system. See: man smb.conf

My buddy’s system was a stock dell, 1tb, 3 partition, no frills system. His kids/grandkids put their games on it (including 20 viruses) and made it unusable. I used clonezilla to “ghost” his /sda1 (tiny, fat32) and /sda2 (14gb recovery), slammed Suse L&G on /sda2 and only had a few of gigs left for pics, music, video; largish stuff. So, I pretty much had to use /sda3, the windows system partition.

Being an email/facebook/wp/spreadsheet workstation with 1 or 2 laptops sharing files wirelessly, and no exposure of Samba to the internet (user authentication only), I don’t think it is a concern (KYFC).

Is there a smarter way to get him access to the 967GB rest of the defunct windoz drive?

AFAIK YaST is not responsible for logging. It is merely a configuration tool. YaST does not put logs anywhere.

I think I meant YAST in the broader sense in that after booting and before YASTing the log is fine. After YAST, it disappeared into oblivion.

Killing the smbds and nmbd and restarting them with the -l <log_dir> option takes mean old yast out. Next time I reboot it, I will try your “system-ng” trick to fix the “system-nFg” problem. :wink:

Thanks 1E6,

BB

P.S. You may want to look in on BunnyMan (Rabich?); he may have gone non-linear. ;(

Perhaps but I’m also having a lot of fun looking at you ramble on - it makes this coffee taste so much better.

I like how you waste so much time on this issue when you could have just dropped it, accepted that you don’t know what you’re doing and learn from it. Instead, you went on another ramble - I bet people at your job, should you have one, really value your personality.

Perhaps but I’m also having a lot of fun looking at you ramble on - it makes this coffee taste so much better.

I like how you waste so much time on this issue when you could have just dropped it, accepted that you don’t know what you’re doing and learn from it. Instead, you went on another ramble - I bet people at your job, should you have one, really value your personality.

RABICH,

You should be checked for rabies. How hypocritical to whine about “how you waste so much time”. YOU??? You are projecting.

This help forum is supposed to be for the exchange of ideas, not pusillanimous, personal attacks. Didn’t your keeper ever teach you that if you can’t say something nice (or helpful), then kindly ST#U.

Do not address any of your vile, woefully under-informed vitriol to me again.

On 2/13/2015 12:56 AM, brianpaustin wrote:
>
<snip>

>
> Killing the smbds and nmbd and restarting them with the -l <log_dir>
> option takes mean old yast out. Next time I reboot it, I will try your
> “system-ng” trick to fix the “system-nFg” problem. :wink:
>
>

That was a typo on my part. “system-ng” should read “syslog-ng”. My apologies. Mea Culpa.


P.V.
“We’re all in this together, I’m pulling for you” Red Green