+ ceph-devel On Mon, Oct 2, 2017 at 1:18 PM, Two Spirit <twospirit6905@xxxxxxxxx> wrote: >>Yes. This is zone to zone sync in RGW. >>RGW geo replication was designed speciflicly to handle slow WAN link. >> >>For more details see: > > Thanks for the links. I'll look them over. > >>Each zone is a complete and separates Ceph cluster. > The diagram in the documentation shows 3 monitors in a single Ceph > cluster when replicating zone to zone so that is what is confusing. > The diagram makes one think there is a single ceph cluster for US-West > and US-East. > > My understanding is that for each separate Ceph cluster, mons > determine quorum for their individual cluster, so the diagram should > show 2 sets of 3 mons. > > On Mon, Oct 2, 2017 at 3:09 AM, Orit Wasserman <owasserm@xxxxxxxxxx> wrote: >> On Fri, Sep 29, 2017 at 6:41 PM, Two Spirit <twospirit6905@xxxxxxxxx> wrote: >>> Is it now supported to be able to write to a secondary zone without problems? >>> >> >> yes from Jewel. >> >>> If not supported, is there a timeframe that this (being able to write >>> to secondary zones) would be supported? >>> >>> It also looks like the federated solution does not scale well with >>> multi region all in sync. >>> >> >> We moved to a new implementation in Jewel (because of problems with >> the old federated solution). >> Region is now called zonegroup to avoid confusion with AWS region. >> A zone in RGW is like an AWS region. >> >>> Is there support in Ceph for region-to-region syncing. Based on the >>> figure in the documentation, it appears that zone to zone syncing does >>> not happen across a slow wan link (as there are 3 MONs in the backend, >>> and unified Ceph cluster is not recommended across a WAN) >> >> Yes. This is zone to zone sync in RGW. >> RGW geo replication was designed speciflicly to handle slow WAN link. >> Each zone is a complete and separates Ceph cluster. >> >> For more details see: >> http://docs.ceph.com/docs/master/radosgw/multisite/ >> http://events.linuxfoundation.org/sites/events/files/slides/Linuxcon_multisitev2.pdf >> >> Regards, >> Orit >> >>> -- >>> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in >>> the body of a message to majordomo@xxxxxxxxxxxxxxx >>> More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
