use partx for CentOS/RHEL instead of partprobe

[alfredodeza]

Those distros will force a user to reboot in order to get the kernel to re-read the partition that was created in the process of creating an OSD.

For a full/detailed description of the issue, see: http://tracker.ceph.com/issues/7334
Fix was verified on CentOS by adding two OSDs and seeing the partitions correctly listed:

[ubuntu@apama002 ~]$ sudo parted /dev/sdc print
Model: HP LOGICAL VOLUME (scsi)
Disk /dev/sdc: 3001GB
Sector size (logical/physical): 512B/512B
Partition Table: gpt

Number  Start   End     Size    File system  Name          Flags
 1      1049kB  5370MB  5369MB               ceph journal
 2      5370MB  10.7GB  5369MB               ceph journal