Ahh, and because that was my first insight into placing object into my ceph pools I (Incorrectly) made some assumptions!
From: Gregory Farnum [mailto:[email protected]] Sent: Friday, 25 September 2015 9:46 AM To: Cory Hawkless <[email protected]> Cc: John Spray <[email protected]>; Ceph Users <[email protected]> Subject: Re: [ceph-users] Basic object storage question On Sep 24, 2015 5:12 PM, "Cory Hawkless" <[email protected]<mailto:[email protected]>> wrote: > > Hi all, thanks for the replies. > So my confusion was because I was using "rados put test.file someobject > testpool" > This command does not seem to split my 'files' into chunks when they are > saved as 'objects', hence the terminology > > Upon bolting openstack Glance onto Ceph I can see hundreds of smaller objects > are created per ISO, this is a much more expected behaviour! > > So does the rados command line application not split files when it is writing > them into Ceph? Correct. The rados cli program is more of an access and admin tool than anything intended for regular use. -Greg > > -----Original Message----- > From: John Spray [mailto:[email protected]<mailto:[email protected]>] > Sent: Thursday, 24 September 2015 6:04 PM > To: Cory Hawkless <[email protected]<mailto:[email protected]>> > Cc: [email protected]<mailto:[email protected]> > Subject: Re: [ceph-users] Basic object storage question > > On Thu, Sep 24, 2015 at 1:51 AM, Cory Hawkless > <[email protected]<mailto:[email protected]>> wrote: > > Hi all, > > > > > > > > I have basic question around how Ceph stores individual objects. > > > > Say I have a pool with a replica size of 3 and I upload a 1GB file to > > this pool. It appears as if this 1GB file gets placed into 3PG’s on 3 > > OSD’s , simple enough? > > Well, you've gone straight from asking about *objects* to talking about > uploading a *file*, so that doesn't make sense :-) > > When you write a file in CephFS, it gets striped into many objects (4MB by > default) in RADOS. Same with objects in RGW, and block devices in RBD. The > stripes are 4MB by default, which results in good data distribution for most > workloads. So the short answer is that unless you're writing directly to > RADOS (i.e with librados), you don't need to worry. > > John > > > Are individual objects never split up? What if I want to storage > > backup files or Openstack Glance images totalling 100’s of GB’s. > > > > > > > > Potentially I could run into issues is I have an object who’s size > > exceeds the available space on any of the OSD’s, say I have 1TB OSD’s > > and they are all 50% full and I try to upload a 501GB image, I presume > > this would fail even through there is sufficient space in the pool > > there is not a single OSD with >500GB of space available. > > > > > > > > Do I have this right? If so is there any way around this? Ideally I’d > > like to use Ceph as target for all of my servers backups, but some of > > these total in the TB’s but none of my OSD’s are this big(Currently > > using 900GB SAS disks). > > > > > > > > Thanks in advance > > > > Cory > > > > > > > > > > > > > > _______________________________________________ > > ceph-users mailing list > > [email protected]<mailto:[email protected]> > > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > _______________________________________________ > ceph-users mailing list > [email protected]<mailto:[email protected]> > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
