bind memory usage

[Peter Dambier]

I can confirm bind 9.4 does run on an (IBM, not Intel) 486-SCL/2 with 16 MB.
That cpu can address no more than 16 MB.

$ cat /proc/meminfo
        total:    used:    free:  shared: buffers:  cached:
Mem:  14540800 10596352  3944448  3194880  1003520  3518464
Swap: 133885952 11907072 121978880
MemTotal:     14200 kB
MemFree:       3852 kB
MemShared:     3120 kB
Buffers:        980 kB
Cached:        3436 kB
SwapTotal:   130748 kB
SwapFree:    119120 kB


So it is consuming 11 MB swap right now.

Bind performs reasonably. It is my local resolver and it is
running an alternative root-zone plus a couple of other zones.

vanadium      up  5+14:18,     1 user,   load 0.00, 0.00, 0.00

It is running for 5 and a half days now.

Freshly started the swapsize is zero but after a day or two
swap is waxing, never waning.

Kind regards
Peter


JINMEI Tatuya / 神明達哉 wrote:
> At Sat, 13 Dec 2008 11:50:52 -0200,
> Leonardo Rodrigues Magalhães <leolistas at solutti.com.br> wrote:
>  
>>     i'm trying to run bind 9.5.0-P2 on a very low memory system. It's a 
>> RouterBoard 450 with 32Mb RAM running OpenWRT.
>>
>> root at sede:~# cat /proc/meminfo
>> MemTotal:        29920 kB
>>
>>     the problem is that bind seems to consume a LOT of memory ... well, 
>> a lot for low memory devices, i never noticed that on machines with BS 
>> of RAM.
> 
> [snip]
> 
>>     question is .... is there something i can do to low bind's memory 
>> usage and successfully run it on those very low embedded devices ???
> 
> Admittedly, BIND9 tends to require a lot of memory.  I'm not sure if
> it can reasonably function with a total system memory of 32MB.
> 
> Some related points:
> - if you enable threads, disable them.  With the thread support BIND9
>   will require even more memory.
> - "max-cache-size 1048576" is a meaningless configuration:
>              Any positive values less than 2MB will be ignored reset
>              to 2MB.
>   (from ARM)
> - 'rndc flush' doesn't release allocated system memory.  It just
>   frees all cache entries within the BIND9 process, so it's not
>   surprising that you didn't see the memory footprint decrease after
>   the flush operation.
> 
> ---
> JINMEI, Tatuya
> Internet Systems Consortium, Inc.
> _______________________________________________
> bind-users mailing list
> bind-users at lists.isc.org
> https://lists.isc.org/mailman/listinfo/bind-users

-- 
Peter and Karin Dambier
Cesidian Root - Radice Cesidiana
Rimbacher Strasse 16
D-69509 Moerlenbach-Bonsweiher
+49(6209)795-816 (Telekom)
+49(6252)750-308 (VoIP: sipgate.de)
mail: peter at peter-dambier.de
http://www.peter-dambier.de/
http://iason.site.voila.fr/
https://sourceforge.net/projects/iason/
ULA= fd80:4ce1:c66a::/48