2018-03-29 11:50 GMT+02:00 David Rabel <rabel@xxxxxxxxxxxxx>:
On 29.03.2018 11:43, Janne Johansson wrote:
> 2018-03-29 11:39 GMT+02:00 David Rabel <rabel@xxxxxxxxxxxxx>:
>
>> For example a replicated pool with size 4: Do i always have to set the
>> min_size to 3? Or is there a way to use min_size 2 and use some other
>> node as a decision maker in case of split brain?
>>
>
> min_size doesn't arbitrate decisions other than
> "can I write if there are only X visible copies?", where X needs to be >
> min_size
> to allow writes.
>
> It doesn't control any logic, it controls the risk level you want to take.
You are right. But with my above example: If I have min_size 2 and size
4, and because of a network issue the 4 OSDs are split into 2 and 2, is
it possible that I have write operations on both sides and therefore
have inconsistent data?
You always write to the primary, which in turn sends copies to the 3 others,
so in the 2+2 split case, only one side can talk to the primary OSD for that pg,
so writes will just happen on one side at most.
May the most significant bit of your life be positive.
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
