List Info

Thread: Re: max-cache-size doesn't work with 9.5.0b1




Re: max-cache-size doesn't work with 9.5.0b1
country flaguser name
Hungary
2008-02-13 06:06:58
On 02/12/08 18:55, JINMEI Tatuya / 神明達哉 wrote:
> At Tue, 12 Feb 2008 18:10:12 +0100,
> Attila Nagy <brafsn.hu> wrote:
>
>   
>>> Then named will listen on
[your_ip_address]:some_port, and you can
>>> browse internal statistics by accessing
>>> http://[your_ip_address]:some_port with your
browser.  When you notice
>>> the memory starts growing, retrieving the
information several times,
>>> and compare the "Memory" section at
the end of the page.  If the
>>> memory hog is inside named, there should be
significant growth in some
>>> of the rows accordingly.
>>>   
>>>       
>> Here are the results:
>> http://people.fsn.hu/~bra/freebsd/bind950-memory-20080
212/
>> and at the end, the relevant line from top.
>>
>> Is this FreeBSD leaking then?
>>     
>
> Looking at the last stat
> http://people.fsn.hu/~bra/freebsd/bind950-memory-
20080212/bind4
> even the total of "MaxUse" is about 69MB,
while the ps output
> indicates the resident size is 441MB.  So, yes, there
should be
> something odd not directly related to the named's
(normal) behavior.
>
> Can you try the same test with only one worker thread
(by using the
> '-n 1' command line option)?  Then we may be able to
chase the problem
> further.
>   
Of course. See the bindn1 files at the same location. (only
the memory 
section included)
The effect is pretty much the same.

-- 
Attila Nagy                                   e-mail:
Attila.Nagyfsn.hu
Free Software Network (FSN.HU)                 phone: +3630
306 6758
http://www.fsn.hu/

_______________________________________________
freebsd-performancefreebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-p
erformance
To unsubscribe, send any mail to
"freebsd-performance-unsubscribefreebsd.org"

Re: max-cache-size doesn't work with 9.5.0b1
country flaguser name
United States
2008-02-13 14:44:01
At Wed, 13 Feb 2008 13:06:58 +0100,
Attila Nagy <brafsn.hu> wrote:

> Of course. See the bindn1 files at the same location.
(only the memory 
> section included)
> The effect is pretty much the same.

Okay, then please try this patch with '-n 1' (note: this
patch doesn't
contain the memory statistics hack via the HTTP interface,
but I don't
we don't need it for this test).

---
JINMEI, Tatuya
Internet Systems Consortium, Inc.


_______________________________________________
freebsd-performancefreebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-p
erformance
To unsubscribe, send any mail to
"freebsd-performance-unsubscribefreebsd.org"
  
[1-2]

about | contact  Other archives ( Real Estate discussion Medical topics )