[Lustre-discuss] "Setstripe count -1" what's the happen when one of the osts outs of space disk?

[Đào Thị Thảo]

oho, I see.
Thank Andreas & Kevin!


On Tue, Apr 6, 2010 at 12:44 PM, Kevin Van Maren <Kevin.Vanmaren at sun.com>wrote:

> Just to clarify,
>
> I believe Andreas is talking about creating the file after the OST is
> already known to be out of space.
>
> If a file is already striped across an OST, when that OST fills up, writes
> that map to that OST will fail
>
> if they require additional storage (ie, overwrites will work, but growing
> the file will return ENOSPC
> once you hit the full OST).
>
> Kevin
>
>
> Andreas Dilger wrote:
>
>> On 2010-04-05, at 02:43, Đào Thị Thảo wrote:
>>
>>
>>> I'm using lustre 1.8.0 on Centos 5.0.
>>> I have a question as follows: I setstripe file for all osts with
>>>  parameter count -1, what's the happen when one of the osts outs of  space
>>> disk?
>>>
>>>
>>
>>
>> If you specify stripe_count=-1 it will stripe across all "available"
>>  OSTs, which should skip the full OST.  If you specify stripe_count=N,
>>  where N = number of OSTs (including the full one) then the allocation  will
>> fail.
>>
>> Cheers, Andreas
>> --
>> Andreas Dilger
>> Principal Engineer, Lustre Group
>> Oracle Corporation Canada Inc.
>>
>> _______________________________________________
>> Lustre-discuss mailing list
>> Lustre-discuss at lists.lustre.org
>> http://lists.lustre.org/mailman/listinfo/lustre-discuss
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20100407/1ca72846/attachment.htm>