I think he needs the “gateway” servers because he wishes to expose the storage to clients which won’t speak Ceph natively. I’m not sure I would entirely trust that windows port of CephFS and there are also security concerns with allowing end users to talk directly to Ceph. There’s also future stuff like CephFS snapshots, which Samba can then wrap around to provide previous versions directly into windows explorer.
You probably also don’t want to use CephFS at this point, there are several outstanding points that means that it’s going to require a bit more work than you ideally want to have to go through if you want to export it HA over NFS and SMB reliably. There were some posts about a year ago (I think??) which covered some of the issues using CTDB with CephFS in a HA fashion. Mainly around timeouts in CephFS and how they match up to what CTDB expects. Currently the simplest and most reliable way is like the blog article linked, to export NFS+SMB on a filesystem over a RBD. This will require the use of Pacemaker for HA and that’s a whole separate topic in itself to get right. From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of David Turner Sent: 20 January 2017 19:08 To: John Petrini <jpetr...@coredial.com>; Joao Eduardo Luis <j...@suse.de> Cc: ceph-users@lists.ceph.com; hen shmuel <hens...@gmail.com> Subject: Re: [ceph-users] [Ceph-community] Consultation about ceph storage cluster architecture CephFS does not require a central NFS server. Any Linux server can mount the CephFS volume at the same time. There is also a windows client for CephFS (https://drupal.star.bnl.gov/STAR/blog/mpoat/cephfs-client-windows-based-dokan-060). I don't see the need for the NFS/SMB server or complicated HA setup. What you do need to make sure is HA is the MDS service. It is not natively HA, but there are several guides to work around that. _____ <https://storagecraft.com> David Turner | Cloud Operations Engineer | <https://storagecraft.com> StorageCraft Technology Corporation 380 Data Drive Suite 300 | Draper | Utah | 84020 Office: 801.871.2760 | Mobile: 385.224.2943 _____ If you are not the intended recipient of this message or received it erroneously, please notify the sender and delete it, together with any attachments, and be advised that any dissemination or copying of this message is prohibited. _____ _____ From: ceph-users [ceph-users-boun...@lists.ceph.com] on behalf of John Petrini [jpetr...@coredial.com] Sent: Friday, January 20, 2017 10:41 AM To: Joao Eduardo Luis Cc: ceph-users@lists.ceph.com; hen shmuel Subject: Re: [ceph-users] [Ceph-community] Consultation about ceph storage cluster architecture Here's a really good write up on how to cluster NFS servers backed by RBD volumes. It could be adapted to use CephFS with relative ease. https://www.sebastien-han.fr/blog/2012/07/06/nfs-over-rbd/ ___ John Petrini NOC Systems Administrator // CoreDial, LLC // <http://coredial.com/> coredial.com // <https://twitter.com/coredial> <http://www.linkedin.com/company/99631> <https://plus.google.com/104062177220750809525/posts> <http://success.coredial.com/blog> Hillcrest I, 751 Arbor Way, Suite 150, Blue Bell PA, 19422 P: 215.297.4400 x232 // F: 215.297.4401 // E: <mailto:jpetr...@coredial.com> jpetr...@coredial.com <http://cta-redirect.hubspot.com/cta/redirect/210539/4c492538-6e4b-445e-9480-bef676787085> The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. On Fri, Jan 20, 2017 at 12:35 PM, Joao Eduardo Luis <j...@suse.de <mailto:j...@suse.de> > wrote: Hi, This email is better suited for the 'ceph-users' list (CC'ed). You'll likely find more answers there. -Joao On 01/20/2017 04:33 PM, hen shmuel wrote: im new to Ceph and i want to build ceph storage cluster at my work site, to provide NAS services to are clients, as NFS to are linux servers clients, and as CIFS to are windows servers clients, to my understanding in order to do that with ceph i need to: 1. build a full ceph storage cluster 2. create CephFS "volumes" on my ceph cluster 3. mount the CephFS to a linux server that will be used as a "gateway" for export the CephFS as NFS to linux servers and as CIFS to windows servers 4. on my linux "gateway" server i need to install NFS server and SMB Server to do the export part 5. in order to overcome the "single point of failure" with this linux "gateway" server in will need to build a cluster of them and use clustered NFS Server and CTBD or something like that i wanted to know if i understand it properly and if this is the right way to do that or if there is a simplified way to achieved NAS services like i want. thanks for any help! _______________________________________________ Ceph-community mailing list ceph-commun...@lists.ceph.com <mailto:ceph-commun...@lists.ceph.com> http://lists.ceph.com/listinfo.cgi/ceph-community-ceph.com _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com <mailto: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
