Hi Gilles, On Wed, Sep 11, 2024 at 10:33 PM Gilles Mocellin < gilles.mocellin@xxxxxxxxxxxxxx> wrote: > Yes, I've read some stories about that in this mailing list, but, my > question > was not clear enough, I want to know what's possible to do with cephadm : > - Is it possible to create several Ingress with cephadm ? (I can't reach > my > test env to test until tomorrow !) > Yes, it is possible, each ingress service will manage a different set VIPs/Subnets, if by any chance they are on the same subnet, you need to set different first_virtual_router_id > - If we do, how can we configure differently thoses groups of RGWs daemons > ? > (related to my previous question on ceph config / ceph.conf) > You can define different groups of RGWs in Cephadm. Each RGW group would be assigned to a specific Ingress service, one for public client requests and the other for RGW multisite replication. Then, on the public RGW group, you would set rgw_run_sync_thread false. Regards. > > Le mercredi 11 septembre 2024, 22:23:14 CEST Anthony D'Atri a écrit : > > I've seen deployments with more than one cohort of RGWs. E.g. > > > > * one group of RGWs that serve internal requests with an LB > > * one group of RGWs that serve external requests with a different LB > > * one group of RGWs that only process LC / GC events > > > > > On Sep 11, 2024, at 4:01 PM, Gilles Mocellin > > > <gilles.mocellin@xxxxxxxxxxxxxx> wrote: > > > > > > Hi again, > > > > > > I wonder if there is a better way of configuring multisite > configuration, > > > especially zone/zonegroup endpoints : > > > > > > - many endpoints, pointing directly at RGW daemons > > > - one endpoint, pointing at the load balancer VIP of Ingress > > > > > > If we want to separate client access from replication, is it possible > to > > > have two Ingress ? It can be great to have them in two different > > > networks. > > > > > > _______________________________________________ > > > ceph-users mailing list -- ceph-users@xxxxxxx > > > To unsubscribe send an email to ceph-users-leave@xxxxxxx > > > > _______________________________________________ > ceph-users mailing list -- ceph-users@xxxxxxx > To unsubscribe send an email to ceph-users-leave@xxxxxxx > _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |