Hello, If you are interested in this topic, we can discuss it in the weekly meeting, but I don't know whether the meeting time works for you or not.
http://webchat.freenode.net/?channels=opnfv-meeting 8:00-9:00 UTC (During winter time, means CET 9:00 AM) Best Regards Chaoyi Huang (joehuang) ________________________________ From: Klemen Pogacnik [kle...@psi-net.si] Sent: 09 January 2017 20:57 To: joehuang Cc: opnfv-tech-discuss@lists.opnfv.org Subject: Re: [opnfv-tech-discuss] [multisite] Multisite VNF Geo site disaster recovery Thanks for answer. If I understand right, currently there is no work on DR with 2 or more Openstack clouds. Is it maybe planned for the future? Kemo On Mon, Jan 9, 2017 at 7:50 AM, joehuang <joehu...@huawei.com<mailto:joehu...@huawei.com>> wrote: Hello, Thank you for the mail, please see inline comments: Best Regards Chaoyi Huang (joehuang) ________________________________ From: opnfv-tech-discuss-boun...@lists.opnfv.org<mailto:opnfv-tech-discuss-boun...@lists.opnfv.org> [opnfv-tech-discuss-boun...@lists.opnfv.org<mailto:opnfv-tech-discuss-boun...@lists.opnfv.org>] on behalf of Klemen Pogacnik [kle...@psi-net.si<mailto:kle...@psi-net.si>] Sent: 06 January 2017 22:05 To: opnfv-tech-discuss@lists.opnfv.org<mailto:opnfv-tech-discuss@lists.opnfv.org> Subject: [opnfv-tech-discuss] [multisite] Multisite VNF Geo site disaster recovery I'm trying to deploy Multisite VNF Geo site disaster recovery scenario with Ceph as a backend for a Cinder service (scenario 3, 3rd way) . Ceph rbd mirroring is used for data replication between two sites. Volume created on site1 is replicated to site2, but this volume is not visible on Cinder on side2. Probably some metadata from Cinder DB must be replicated too. Has anybody been thinkink or even working on that? ---> yes, current replication implementation will hide all these information from the API level, and the replication fail-over, fail-back are activities inside one OpenStack domain. you have to replicate the db to the site2 if you want to do DR with different OpenStack cloud. We had discussed with the community, whether it's reasonable to expose the volume reference from the admin API, community prefer to let the driver to deal with these information. The second question is, how to make switchover, when site one goes down. On open stack there is cinder failover-host command, which is, as I can understand, only useful for configuration with one openstack and two ceph clusters. Any idea how to make switchover with two openstacks each connected to its own ceph cluster. Thanks a lot for help! ---> This needs to be triggered by GR software, or manually, or via DNS(if applicable). Kemo
_______________________________________________ opnfv-tech-discuss mailing list opnfv-tech-discuss@lists.opnfv.org https://lists.opnfv.org/mailman/listinfo/opnfv-tech-discuss
