samba rename bug

[Eric Shubert]

Craig White wrote:
> On Mon, 2009-08-17 at 18:29 -0700, Eric Shubert wrote:
>> Back in March I created this bug report: 
>> https://bugzilla.samba.org/show_bug.cgi?id=6173
>>
>> That was on samba-3.0.33, which is the current version on CentOS5. I was 
>> notified today that 3.0 is no longer under maintenance. As it turns out, 
>> 3.2.x (the version in FC10) is also out of maintenance. The current 
>> stable release of samba is 3.4.0, released 7/3/09. FC11 is on 3.3.x. I'm 
>> getting a clear picture of who's the tortoise and who's the hare. ;) I 
>> expect that FC12 will have samba-3.4.x.
>>
>> Anyway, I'm wondering if by chance anyone's running 3.3 or 3.4 and can 
>> verify whether this bug still exists or not. If it does, I had a patch 
>> for the 3.0 source that I'll look into reworking and submitting to the 
>> 3.4 (preferrably) or 3.3 branch.
>>
>> The bug is really easy to test. See the bug report. I just don't have a 
>> recent version of samba operational yet, and don't want to waste any 
>> time on it if the bug's been fixed.
> ----
> I'm pretty sure that sernet.de has rpm's for RHEL/CentOS v5 (Volker is
> Sernet.de) that you can install without too much effort but yes, RHEL 5
> is still backporting security patches only to 3.0.33. Didn't that patch
> come from Volker?

I've no idea where 3.0.33 came from.
Thanks for the Volker reference. Have you used his packages at all? His 
site looks pretty professional, so I probably shouldn't be concerned.

Now all I need to do is figure out how to migrate from 3.0 to 3.3 
smoothly. Any insight?

> No, I've not installed any version other than Fedora 10/11 and
> RHEL/CentOS versions to test

Do you have 3.3 on FC11 you can do a test with? It's really simple. On a 
windows machine with N: (or whatever drive) mapped to a samba share, 
open a command prompt and execute the following commands:
N:\>echo test > testfile
N:\>attrib +r testfile
N:\>attrib testfile
A    R     N:\testfile
N:\>ren testfile testren
Access is denied.

If you don't get the last reply, then the bug's fixed. If you get the 
denied message, then the bug lives on.

-- 
-Eric 'shubes'