On 12/24/2011 1:04 PM, gene heskett wrote:
> On Saturday, December 24, 2011 12:56:52 PM Mark Wendt (Contractor) did
> opine:
>
>    
>> On 12/24/2011 12:22 PM, gene heskett wrote:
>>      
>>> On Saturday, December 24, 2011 12:14:41 PM yann jautard did opine:
>>>        
>>>> Le 24/12/2011 15:04, gene heskett a écrit :
>>>>          
>>>>> On Saturday, December 24, 2011 09:00:31 AM Mark Wendt (Contractor)
>>>>> did
>>>>>
>>>>> opine:
>>>>>            
>>>>>> On 12/23/2011 2:47 PM, gene heskett wrote:
>>>>>>              
>>>>>>> I sounded like a good idea, but:
>>>>>>> [gene@coyote ~]$ ssh shop
>>>>>>> gene@shop's password:
>>>>>>> Linux shop 2.6.32-122-rtai #rtai SMP Tue Jul 27 12:44:07 CDT 2010
>>>>>>> i686 GNU/Linux
>>>>>>> Ubuntu 10.04.3 LTS
>>>>>>>
>>>>>>> Welcome to Ubuntu!
>>>>>>>
>>>>>>>      * Documentation:  https://help.ubuntu.com/
>>>>>>>
>>>>>>> 11 packages can be updated.
>>>>>>> 6 updates are security updates.
>>>>>>>
>>>>>>> Last login: Thu Dec 22 09:38:52 2011 from coyote.coyote.den
>>>>>>> gene@shop:~$ sudo useradd -u 500 gene
>>>>>>> [sudo] password for gene:
>>>>>>> useradd: user 'gene' already exists
>>>>>>>
>>>>>>> So there isn't an obvious way to make the user numbers match
>>>>>>> between the *buntu's and the rest of the world.
>>>>>>>
>>>>>>> The last time I tried that, I wound up re-installing to fix it.
>>>>>>>
>>>>>>> Cheers, Gene
>>>>>>>                
>>>>>> Gene,
>>>>>>
>>>>>> What about good old vi, or gedit on the /etc/passwd and /etc/group
>>>>>> files, changing the uid and gid to what ever you need, then doing a
>>>>>> chown -R gene:gene on /home/gene
>>>>>>
>>>>>> No need to reinstall.  Just a little careful editing is all you
>>>>>> need.
>>>>>>
>>>>>> Mark
>>>>>>              
>>>>> I did something like that, including the chown -R back on 8.04 and
>>>>> had to reinstall.  Among other things, sudo quit working so I
>>>>> couldn't fix the rest of the perms problems that created.
>>>>>
>>>>> Cheers, Gene
>>>>>            
>>>> yeah sudo quit working due to permission problems during the
>>>> operation.
>>>>
>>>> This is why you need to create a root password first, and login as
>>>> root to make the user modification.
>>>>
>>>> sudo password root
>>>>
>>>> then you log off the graphical interface
>>>>
>>>> switch to terminal (ctrl-F1)
>>>>
>>>> login as root
>>>>
>>>> make the modifications
>>>>
>>>>
>>>> go back to the graphical login (ctrl-F7 or F8) then login as your
>>>> normal user, and that's all.
>>>>          
>>> That is, IIRC, what I did to an older 6.06 LTS install.  Things worked
>>> passably well, but somehow the root passwords presence messed up sudo,
>>> it wouldn't take either pw, so that I had to constantly su - to do
>>> things that scripts use su for.  So I tried to remove the root pw,
>>> then that blew everything up and I had to re-install.
>>>
>>> AFAIAC, the buntu's do that to be a PITA, thinking it might add to the
>>> many layers of security.  Perhaps it does, to an ex winders user, but
>>> I am used to machinery that only I have access to, and which do
>>> exactly as I tell them too, even if its wrong. :)
>>>
>>> Cheers, Gene
>>>        
>> Gene,
>>
>> That sounds like syntax problems in the passwd, group or shadow file.
>> The root account's password has nothing to do with the operation of
>> sudo.  sudo uses either a set uid, or set gid process to gain the
>> elevated privileges to do it's work.  It doesn't access the root account
>> at all.
>>
>> Realize there's a difference between a simple "su" and  "su -".  An "su"
>> will bring you up to superuser, however it uses the rc scripts in the
>> account you are "su'ing" from to set the environment.  An "su -" brings
>> you up to superuser, but it does so using the rc scripts in the "root"
>> account to set the environment.  Unless you have a reason to use the
>> regular user account's rc scripts, I'd recommend to always use "su -"
>> when you are doing real superuser work.
>>
>> Mark
>>      
> I do.  But that is so all encompassing on pclos, that all paths then have
> to be cd'd to from the /root account.  Even when using it in a script, a cd
> to do something in a subdir must be semicolon separated else the effect of
> the cd expires at the end of the current line of the script, so the
> operative work command must be "cd wherever;exec the subscript" in
> construction.  You cannot cd somewhere, and expect that cd to be effective
> for the next line of the script, it is not.  One can script around it, but
> it took me a half an hour to grasp the concept.  It will be interesting to
> see if centos has a similar restriction.
>
> Cheers, Gene
>    
Or just run the script with the entire path: 
/run/this/script/in/this/directory/script

Mark


------------------------------------------------------------------------------
Write once. Port to many.
Get the SDK and tools to simplify cross-platform app development. Create 
new or port existing apps to sell to consumers worldwide. Explore the 
Intel AppUpSM program developer opportunity. appdeveloper.intel.com/join
http://p.sf.net/sfu/intel-appdev
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to