The `rados -p <pool> df` doesn't account for replication? So if it says I have 10TB utilized, I am actually storing 5TB of data? Or in other words, if I have 6TB free, I can only store 3TB more?
As I was saying, in regards to Mark's comments, after the image is imported, the difference in utilization reported by `rados -p <pool> df` before and after the import is the same size as the image just imported. Here, during the import process, `rados -p <pool> df` is reporting 2x disk utilization. But from what Mark reports and what I have also seen, I should expect disk utilization to decrease from 2x to 1x. I was curious about this, and then saw this thread discussing it. Note, I am not looking at the results of 'df /ceph/mounted/volume' from the file system. I am looking at rados's df output on the pool. If I was comparing the reports of the file system df output, then I would expect 2x disk utilization, which accounts for replication. -RG ----- Original Message ----- From: "Udo Lembke" <ulem...@polarzone.de> To: ceph-users@lists.ceph.com Sent: Thursday, April 10, 2014 1:35:51 PM Subject: Re: [ceph-users] OSD space usage 2x object size after rados put Hi, On 10.04.2014 20:03, Russell E. Glaue wrote: > I am seeing the same thing, and was wondering the same. > > We have 16 OSDs on 4 hosts. The File system is Xfs. The OS is CentOS 6.4. > ceph version 0.72.2 > > I am importing a 3.3TB disk image into a rbd image. > At 2.6TB, and still importing, 5.197TB is used according to `rados -p <pool> > df` that's looks normal for me. With replication of 2 you have all data twice: 2*2.6TB=5.2TB Udo _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
