[Gluster-users] glusterfs and samba (file-max limit reached)

Anand Avati avati at gluster.com
Tue Apr 28 20:00:58 UTC 2009


What happens when you stat a filename, which you know exists on only
one subvolume? does it get healed then atleast?

Avati

On Tue, Apr 28, 2009 at 9:43 PM, Luca Barbato <lu_zero at gentoo.org> wrote:
> Anand Avati wrote:
>>>>
>>>> what is the git commit id of the snapshot you are using?
>>>
>>> I' updated to head, now the fd issue seems gone, autoheal is gone as
>>> well,
>>> apparently...
>>
>> can you please explain what you mean by autoheal being gone as well?
>
> I have a setup with 2 replicate over 3 bricks
>
> each node holds a pair of brick.
>
> a node got offline for outside issues and during the usage a second one got
> offline because of the fd leak.
>
> now everything is online and the content of a directory present just on the
> node that remained online doesn't show up, I don't see any message about
> autohealing being triggered.
>
> same happens with other files that shown up as the older copy present on the
> node being offline the most time.
>
> I'm not sure what is exactly wrong.
>
> --
>
> Luca Barbato
> Gentoo Council Member
> Gentoo/linux Gentoo/PPC
> http://dev.gentoo.org/~lu_zero
>
>




More information about the Gluster-users mailing list