problem with nfs ls latency during high IO
Judith Flo Gaya
jflo at imppc.org
Mon Mar 14 23:12:36 UTC 2011
Thanks a lot for your suggestion, but async looks to me a little "dangerous" don't you think?
Regarding the kernel, I've seen that the new update in fedora's kernel will not change to the stable one.
What do you suggest? Shall I compile the new kernel?
I have to put this server on production soon and I dont't want my clients to get stuck like this ;(
Thanks a lot anyway!
j
On 14/03/2011, at 19:36, Michael Cronenworth <mike at cchtml.com> wrote:
> Judith Flo Gaya wrote:
>> Can you help me on this? am I wrong in my supposition (is the patch
>> really applied)? is it possible that my problem is somewhere else?
>>
>
> The patch looks to be backported (per BZ comment). Meaning: It's already
> upstream. The latest stable kernel (2.6.37) seems to have the patch[1]
> albeit the code is vastly different.
>
> Have you tried setting "async" on the NFS server shares? I have had to
> do that on my systems. Everything gets bogged down in large gigabyte
> transfers if I don't.
>
>
> [1]
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=fs/nfs/inode.c;h=e67e31c734163c6fe6c0eef0d338a2b87ae55763;hb=3c0eee3fe6a3a1c745379547c7e7c904aa64f6d5#l498
> --
> users mailing list
> users at lists.fedoraproject.org
> To unsubscribe or change subscription options:
> https://admin.fedoraproject.org/mailman/listinfo/users
> Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
More information about the users
mailing list