[Gluster-devel] 2.0.0rc4 locks up when used as root

[nicolas prochazka]

hello,
I've a simple AFR configure, two server, two client.
if I do copy of 10G file, glusterfs client process grows and grows , > 5Go
however I just define

 volume io-cache
type performance/io-cache
option cache-size 2048MB             # default is 32MB
option page-size  128KB             #128KB is default option
option cache-timeout 2  # default is 1
subvolumes iothreads
end-volume

so it is glibc problem or a memory leak ?
memory does not scale down.

Nicolas


On Fri, Mar 20, 2009 at 11:35 AM, Gordan Bobic <gordan at bobich.net> wrote:
> On Thu, 19 Mar 2009 12:43:20 -0700, Amar Tumballi <amar at gluster.com> wrote:
>>>
>>> Thanks for the update, I'll try it tonight. Any news on the memory leak?
>>> Have you managed to track it down yet?
>>>
>>
>> We noticed that its some issue with 'arena algorithm' of glibc's memory
>> management. We are working on solving this issue properly. Should have an
>> update soon. As of now, there is no 'leak' observed, but all of the
> issues
>> were related to memory fragmentation problems.
>
> Interesting. I can't quite see how memory fragmentation could be
> responsible for such extreme memory usage growth that eventually uses up
> all RAM+swap, but I'll be happy to test this when I manage to get rc5 to
> boot.
>
> Gordan
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>