[Samba] Samba4 on Ubuntu server

[David Bear]

On 12/29/2014 12:52 AM, Andrew Bartlett wrote:
> On Mon, 2014-12-22 at 14:33 -0700, Cj Tibbetts wrote:
>> Here is the information requested
>> it took 3 days before results came
>> samba-tool dbcheck finished with this error
>>
>> root at ham1:/usr/local/samba/bin# time ./samba-tool dbcheck --cross-ncs
>> Checking 212687 objects
>> ERROR(<type 'exceptions.MemoryError'>): uncaught exception -
>>    File
>> "/usr/local/samba/lib/python2.7/site-packages/samba/netcmd/__init__.py",
>> line 175, in _run
>>      return self.run(*args, **kwargs)
>>    File
>> Segmentation fault (core dumped)
>>
>> real    1980m21.023s
>> user    1408m48.304s
>> sys     11m43.788s
> 212000 objects is very large for Samba, but running out of memory isn't
> a good idea either.  We need to work on dbcheck to make it able to
> continue working on such large databases, rather than blowing up.
I don't know why we should have so many objects. We only have about 125 
machines joined to the the AD.

We have about 800 persons objects.

Something is indeed amis.
>
> I'm assuming you have an exceedingly large DNS partition, due to the bug
> in Samba 4.1 where we would create a new deleted object each time DNS
> was updated.
Are you suggesting that if we went with bind_dlz we might have better 
luck ? Or is there regular 'clean up' that we should be doing to remove 
old dns objects?

>
> If a core file was really dumped, getting that into a debugger and
> getting a bt full may be enlightening.
Not sure at this point I can get a core dump. In a separate post to the 
list, I asked a question related to the AD sites and services tool. I'm 
wondering if that tool could have messed things up?
>
> Andrew Bartlett
>


-- 
David Bear
602-903-6476