Re: geo replication
Is there any more information on an ETA? -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RBD
Hi Slawomir, I'm not a ceph specialist, nor a developer, but I think Rados Object Store API and Eleanor Cawthon's paper could be a possible solution for radosgw replication. Automatic recovery for RDB would be impraticable due the size of the clusters. Some databases fixed it by giving a global id to transactions but I believe it would break some ceph rules. If you look at Amazon, they replicate databases using the database technology, not by replicating the storage. If ceph creates a transaction log and the internet goes down for few days, you would have to be able to save all the transactions until it comes back, and then you will have to be able to catch up. But for your radosgw I believe it is possible to reproduce an efficient transaction log by moving the logic and computation from your embedded perl to the librados API (I'm sure if it is correct, I mean the one you put some logic inside the OSDs) to populate a list of transactions stored inside ceph, as described here: http://ceph.com/papers/CawthonKeyValueStore.pdf , it may reduce the sysadmin mistakes you mentioned. The problem is perl, nginx and AMPQ is much simpler than rados and C. If the replication stales the key-value list reduces the replication because it aggregates updated objects with their last state, it also makes it easy to deal with deleted objects, parallel copies and buckets prioritization. If the replica data-center serves read-only requests, adding a little more complexity, it would be possible to replicate objects on demand by checking the transaction log before serving an object, until the replication reaches a certain level of acceptable delay. For a complete data-center recovery, it would be nice to have tools to simplify some operations, for example you could get from the crush map one server of each branch, move them to the lost datacenter and set them all as primary, replicate the data and wait for a recovery from the journals. It is a huge operation that makes sense for a lot of companies and I know some that did something similar for big raid systems. For lots of users like me, replication and its risks would be a valuable and manageable feature and maybe it could be another project, less strict with the fundamentals of ceph. On Wed, Feb 20, 2013 at 2:19 PM, Sage Weil s...@inktank.com wrote: On Wed, 20 Feb 2013, S?awomir Skowron wrote: Like i say, yes. Now it is only option, to migrate data from one cluster to other, and now it must be enough, with some auto features. But is there any timeline, or any brainstorming in ceph internal meetings, about any possible replication in block level, or something like that ?? I would like to get this in for cuttlefish (0.61). See #4207 for the underlying rados bits. We also need to settle the file format discussion; any input there would be appreciated! sage On 20 lut 2013, at 17:33, Sage Weil s...@inktank.com wrote: On Wed, 20 Feb 2013, S?awomir Skowron wrote: My requirement is to have full disaster recovery, buisness continuity, and failover of automatet services on second Datacenter, and not on same ceph cluster. Datacenters have 10GE dedicated link, for communication, and there is option to expand cluster into two DataCenters, but it is not what i mean. There are advantages of this option like fast snapshots, and fast switch of services, but there are some problems. When we talk about disaster recovery i mean that whole storage cluster have problems, not only services at top of storage. I am thinking about bug, or mistake of admin, that makes cluster not accessible in any copy, or a upgrade that makes data corruption, or upgrade that is disruptive for services - auto failover services into another DC, before upgrade cluster. If cluster have a solution to replicate data in rbd images to next cluster, than, only data are migrated, and when disaster comes, than there is no need to work on last imported snapshot (there can be constantly imported snapshot with minutes, or hour, before last production), but work on data from now. And when we have automated solution to recover DB (one of app service on top of rbd) clusters in new datacenter infrastructure, than we have a real disaster recovery solution. That's why we made, a s3 api layer synchronization to another DC, and Amazon, and only RBD is left. Have you read the thread from Jens last week, 'snapshot, clone and mount a VM-Image'? Would this type of capability capture you're requirements? sage Dnia 19 lut 2013 o godz. 10:23 S?bastien Han han.sebast...@gmail.com napisa?(a): Hi, For of all, I have some questions about your setup: * What are your requirements? * Are the DCs far from each others? If they are reasonably close to each others, you can setup a single cluster, with replicas across both DCs and manage the RBD devices with pacemaker.
Re: Geo-replication with RBD
Like i say, yes. Now it is only option, to migrate data from one cluster to other, and now it must be enough, with some auto features. But is there any timeline, or any brainstorming in ceph internal meetings, about any possible replication in block level, or something like that ?? On 20 lut 2013, at 17:33, Sage Weil s...@inktank.com wrote: On Wed, 20 Feb 2013, S?awomir Skowron wrote: My requirement is to have full disaster recovery, buisness continuity, and failover of automatet services on second Datacenter, and not on same ceph cluster. Datacenters have 10GE dedicated link, for communication, and there is option to expand cluster into two DataCenters, but it is not what i mean. There are advantages of this option like fast snapshots, and fast switch of services, but there are some problems. When we talk about disaster recovery i mean that whole storage cluster have problems, not only services at top of storage. I am thinking about bug, or mistake of admin, that makes cluster not accessible in any copy, or a upgrade that makes data corruption, or upgrade that is disruptive for services - auto failover services into another DC, before upgrade cluster. If cluster have a solution to replicate data in rbd images to next cluster, than, only data are migrated, and when disaster comes, than there is no need to work on last imported snapshot (there can be constantly imported snapshot with minutes, or hour, before last production), but work on data from now. And when we have automated solution to recover DB (one of app service on top of rbd) clusters in new datacenter infrastructure, than we have a real disaster recovery solution. That's why we made, a s3 api layer synchronization to another DC, and Amazon, and only RBD is left. Have you read the thread from Jens last week, 'snapshot, clone and mount a VM-Image'? Would this type of capability capture you're requirements? sage Dnia 19 lut 2013 o godz. 10:23 S?bastien Han han.sebast...@gmail.com napisa?(a): Hi, For of all, I have some questions about your setup: * What are your requirements? * Are the DCs far from each others? If they are reasonably close to each others, you can setup a single cluster, with replicas across both DCs and manage the RBD devices with pacemaker. Cheers. -- Regards, S?bastien Han. On Mon, Feb 18, 2013 at 3:20 PM, S?awomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Mon, Feb 18, 2013 at 3:07 PM, S?awomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Fri, Feb 1, 2013 at 12:27 AM, Neil Levine neil.lev...@inktank.com wrote: Skowron, Can you go into a bit more detail on your specific use-case? What type of data are you storing in rbd (type, volume)? Neil On Wed, Jan 30, 2013 at 10:42 PM, Skowron S?awomir slawomir.skow...@grupaonet.pl wrote: I make new thread, because i think it's a diffrent case. We have managed async geo-replication of s3 service, beetwen two ceph clusters in two DC's, and to amazon s3 as third. All this via s3 API. I love to see native RGW geo-replication with described features in another thread. There is another case. What about RBD replication ?? It's much more complicated, and for disaster recovery much more important, just like in enterprise storage arrays. One cluster in two DC's, not solving problem, because we need security in data consistency, and isolation. Do you thinking about this case ?? Regards Slawomir Skowron-- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org 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 majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- - Pozdrawiam S?awek sZiBis Skowron -- - Pozdrawiam S?awek sZiBis Skowron -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this
Re: Geo-replication with RBD
On Wed, 20 Feb 2013, S?awomir Skowron wrote: Like i say, yes. Now it is only option, to migrate data from one cluster to other, and now it must be enough, with some auto features. But is there any timeline, or any brainstorming in ceph internal meetings, about any possible replication in block level, or something like that ?? I would like to get this in for cuttlefish (0.61). See #4207 for the underlying rados bits. We also need to settle the file format discussion; any input there would be appreciated! sage On 20 lut 2013, at 17:33, Sage Weil s...@inktank.com wrote: On Wed, 20 Feb 2013, S?awomir Skowron wrote: My requirement is to have full disaster recovery, buisness continuity, and failover of automatet services on second Datacenter, and not on same ceph cluster. Datacenters have 10GE dedicated link, for communication, and there is option to expand cluster into two DataCenters, but it is not what i mean. There are advantages of this option like fast snapshots, and fast switch of services, but there are some problems. When we talk about disaster recovery i mean that whole storage cluster have problems, not only services at top of storage. I am thinking about bug, or mistake of admin, that makes cluster not accessible in any copy, or a upgrade that makes data corruption, or upgrade that is disruptive for services - auto failover services into another DC, before upgrade cluster. If cluster have a solution to replicate data in rbd images to next cluster, than, only data are migrated, and when disaster comes, than there is no need to work on last imported snapshot (there can be constantly imported snapshot with minutes, or hour, before last production), but work on data from now. And when we have automated solution to recover DB (one of app service on top of rbd) clusters in new datacenter infrastructure, than we have a real disaster recovery solution. That's why we made, a s3 api layer synchronization to another DC, and Amazon, and only RBD is left. Have you read the thread from Jens last week, 'snapshot, clone and mount a VM-Image'? Would this type of capability capture you're requirements? sage Dnia 19 lut 2013 o godz. 10:23 S?bastien Han han.sebast...@gmail.com napisa?(a): Hi, For of all, I have some questions about your setup: * What are your requirements? * Are the DCs far from each others? If they are reasonably close to each others, you can setup a single cluster, with replicas across both DCs and manage the RBD devices with pacemaker. Cheers. -- Regards, S?bastien Han. On Mon, Feb 18, 2013 at 3:20 PM, S?awomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Mon, Feb 18, 2013 at 3:07 PM, S?awomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Fri, Feb 1, 2013 at 12:27 AM, Neil Levine neil.lev...@inktank.com wrote: Skowron, Can you go into a bit more detail on your specific use-case? What type of data are you storing in rbd (type, volume)? Neil On Wed, Jan 30, 2013 at 10:42 PM, Skowron S?awomir slawomir.skow...@grupaonet.pl wrote: I make new thread, because i think it's a diffrent case. We have managed async geo-replication of s3 service, beetwen two ceph clusters in two DC's, and to amazon s3 as third. All this via s3 API. I love to see native RGW geo-replication with described features in another thread. There is another case. What about RBD replication ?? It's much more complicated, and for disaster recovery much more important, just like in enterprise storage arrays. One cluster in two DC's, not solving problem, because we need security in data consistency, and isolation. Do you thinking about this case ?? Regards Slawomir Skowron-- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org 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 majord...@vger.kernel.org More majordomo
Re: Geo-replication with RBD
Hi, For of all, I have some questions about your setup: * What are your requirements? * Are the DCs far from each others? If they are reasonably close to each others, you can setup a single cluster, with replicas across both DCs and manage the RBD devices with pacemaker. Cheers. -- Regards, Sébastien Han. On Mon, Feb 18, 2013 at 3:20 PM, Sławomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Mon, Feb 18, 2013 at 3:07 PM, Sławomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Fri, Feb 1, 2013 at 12:27 AM, Neil Levine neil.lev...@inktank.com wrote: Skowron, Can you go into a bit more detail on your specific use-case? What type of data are you storing in rbd (type, volume)? Neil On Wed, Jan 30, 2013 at 10:42 PM, Skowron Sławomir slawomir.skow...@grupaonet.pl wrote: I make new thread, because i think it's a diffrent case. We have managed async geo-replication of s3 service, beetwen two ceph clusters in two DC's, and to amazon s3 as third. All this via s3 API. I love to see native RGW geo-replication with described features in another thread. There is another case. What about RBD replication ?? It's much more complicated, and for disaster recovery much more important, just like in enterprise storage arrays. One cluster in two DC's, not solving problem, because we need security in data consistency, and isolation. Do you thinking about this case ?? Regards Slawomir Skowron-- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org 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 majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- - Pozdrawiam Sławek sZiBis Skowron -- - Pozdrawiam Sławek sZiBis Skowron -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org 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 majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RBD
Hi, now i can response, after i was sick. Nginx is compiled with perl/or lua support. Inside nginx configuration is hook, for a perl code, or lua code, as you prefer. This code have a inline functionality. We have testing this from logs, but it's not a good idea.Now in line option, have advantage, because, we can reject PUT's if AMQP is not working, and we don't need to resync, all requests. If it took long, than, wee can't disable queue, and go direct, without AMQP, and resync offline, from logs, by a simple admin tool. This in line functionality, working only on DELETE, PUT, and rest are skipped, Every DELETE, PUT, have a own queue, with own priorities, and custom info in header, for calculating, a time, of synchronization. This nginx functionality, only putting data into queues, and every data, are going into our, S3 (ceph), and Amazon s3, via nginx, with almost same configuration, distributed by puppet. On every DataCenter, we have a bunch of workers, getting data from queues, dedicated for a location, and then, they are getting data syncing, from source to destination. If data can't be get from source, than info is going into error queue, and this queue is re-checked, for some time. I am in middle of writing some article, about this, but my sickness, have slow down this process slightly. On Thu, Jan 31, 2013 at 10:50 AM, Gandalf Corvotempesta gandalf.corvotempe...@gmail.com wrote: 2013/1/31 Sławomir Skowron szi...@gmail.com: We are using nginx, on top of rgw. In nginx we manage to create logic, for using a AMQP, and async operations via queues. Then workers, on every side getiing data from own queue, and then coping data from source, to destination in s3 API. Works for PUT/DELETE, and work automatic when production goes on another location. I don't know much about messaging, are you able to share some configuration or more details ? -- - Pozdrawiam Sławek sZiBis Skowron On Thu, Jan 31, 2013 at 10:50 AM, Gandalf Corvotempesta gandalf.corvotempe...@gmail.com wrote: 2013/1/31 Sławomir Skowron szi...@gmail.com: We are using nginx, on top of rgw. In nginx we manage to create logic, for using a AMQP, and async operations via queues. Then workers, on every side getiing data from own queue, and then coping data from source, to destination in s3 API. Works for PUT/DELETE, and work automatic when production goes on another location. I don't know much about messaging, are you able to share some configuration or more details ? -- - Pozdrawiam Sławek sZiBis Skowron -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RBD
Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Mon, Feb 18, 2013 at 3:07 PM, Sławomir Skowron szi...@gmail.com wrote: Hi, Sorry for very late response, but i was sick. Our case is to make a failover rbd instance in another cluster. We are storing block device images, for some services like Database. We need to have a two clusters, synchronized, for a quick failover, if first cluster goes down, or for upgrade with restart, or many other cases. Volumes are in many sizes: 1-500GB external block device for kvm vm, like EBS. On Fri, Feb 1, 2013 at 12:27 AM, Neil Levine neil.lev...@inktank.com wrote: Skowron, Can you go into a bit more detail on your specific use-case? What type of data are you storing in rbd (type, volume)? Neil On Wed, Jan 30, 2013 at 10:42 PM, Skowron Sławomir slawomir.skow...@grupaonet.pl wrote: I make new thread, because i think it's a diffrent case. We have managed async geo-replication of s3 service, beetwen two ceph clusters in two DC's, and to amazon s3 as third. All this via s3 API. I love to see native RGW geo-replication with described features in another thread. There is another case. What about RBD replication ?? It's much more complicated, and for disaster recovery much more important, just like in enterprise storage arrays. One cluster in two DC's, not solving problem, because we need security in data consistency, and isolation. Do you thinking about this case ?? Regards Slawomir Skowron-- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org 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 majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- - Pozdrawiam Sławek sZiBis Skowron -- - Pozdrawiam Sławek sZiBis Skowron -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RBD
We are using nginx, on top of rgw. In nginx we manage to create logic, for using a AMQP, and async operations via queues. Then workers, on every side getiing data from own queue, and then coping data from source, to destination in s3 API. Works for PUT/DELETE, and work automatic when production goes on another location. On Thu, Jan 31, 2013 at 9:25 AM, Gandalf Corvotempesta gandalf.corvotempe...@gmail.com wrote: 2013/1/31 Skowron Sławomir slawomir.skow...@grupaonet.pl: We have managed async geo-replication of s3 service, beetwen two ceph clusters in two DC's, and to amazon s3 as third. All this via s3 API. I love to see native RGW geo-replication with described features in another thread. how did you do this? -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- - Pozdrawiam Sławek sZiBis Skowron -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RBD
Skowron, Can you go into a bit more detail on your specific use-case? What type of data are you storing in rbd (type, volume)? Neil On Wed, Jan 30, 2013 at 10:42 PM, Skowron Sławomir slawomir.skow...@grupaonet.pl wrote: I make new thread, because i think it's a diffrent case. We have managed async geo-replication of s3 service, beetwen two ceph clusters in two DC's, and to amazon s3 as third. All this via s3 API. I love to see native RGW geo-replication with described features in another thread. There is another case. What about RBD replication ?? It's much more complicated, and for disaster recovery much more important, just like in enterprise storage arrays. One cluster in two DC's, not solving problem, because we need security in data consistency, and isolation. Do you thinking about this case ?? Regards Slawomir Skowron-- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org 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 majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RADOS GW
On Jan 28, 2013, at 11:32 AM, Gregory Farnum g...@inktank.com wrote: On Monday, January 28, 2013 at 9:54 AM, Ben Rowland wrote: Hi, I'm considering using Ceph to create a cluster across several data centres, with the strict requirement that writes should go to both DCs. This seems possible by specifying rules in the CRUSH map, with an understood latency hit resulting from purely synchronous writes. The part I'm unsure about is how the RADOS GW fits into this picture. For high availability (and to improve best-case latency on reads), we'd want to run a gateway in each data centre. However, the first paragraph of the following post suggests this is not possible: http://article.gmane.org/gmane.comp.file-systems.ceph.devel/12238 Is there a hard restriction on how many radosgw instances can run across the cluster, or is the point of the above post more about a performance hit? It's talking about the performance hit. Most people can't afford data-center level connectivity between two different buildings. ;) If you did have a Ceph cluster split across two DC (with the bandwidth to support them) this will work fine. There aren't any strict limits on the number of gateways you stick on a cluster, just the scaling costs associated with cache invalidation notifications. It seems to me it should be possible to run more than one radosgw, particularly if each instance communicates with a local OSD which can proxy reads/writes to the primary (which may or may not be DC-local). They aren't going to do this, though — each gateway will communicate with the primaries directly. I don't know what the timeline is, but Yehuda proposed recently the idea of master and slave zones (subsets of a cluster) and other changes to facilitate rgw geo-replication and disaster recovery. See this message: http://article.gmane.org/gmane.comp.file-systems.ceph.devel/12238 If/when that comes to fruition it would open a lot of possibilities for the kind of scenario you're talking about. (Yes, I'm looking forward to it. :) ) JN -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Geo-replication with RADOS GW
On Monday, January 28, 2013 at 9:54 AM, Ben Rowland wrote: Hi, I'm considering using Ceph to create a cluster across several data centres, with the strict requirement that writes should go to both DCs. This seems possible by specifying rules in the CRUSH map, with an understood latency hit resulting from purely synchronous writes. The part I'm unsure about is how the RADOS GW fits into this picture. For high availability (and to improve best-case latency on reads), we'd want to run a gateway in each data centre. However, the first paragraph of the following post suggests this is not possible: http://article.gmane.org/gmane.comp.file-systems.ceph.devel/12238 Is there a hard restriction on how many radosgw instances can run across the cluster, or is the point of the above post more about a performance hit? It's talking about the performance hit. Most people can't afford data-center level connectivity between two different buildings. ;) If you did have a Ceph cluster split across two DC (with the bandwidth to support them) this will work fine. There aren't any strict limits on the number of gateways you stick on a cluster, just the scaling costs associated with cache invalidation notifications. It seems to me it should be possible to run more than one radosgw, particularly if each instance communicates with a local OSD which can proxy reads/writes to the primary (which may or may not be DC-local). They aren't going to do this, though — each gateway will communicate with the primaries directly. -Greg -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: geo replication
On Wed, Jan 9, 2013 at 1:33 PM, Gandalf Corvotempesta gandalf.corvotempe...@gmail.com wrote: 2013/1/9 Mark Kampe mark.ka...@inktank.com: Asynchronous RADOS replication is definitely on our list, but more complex and farther out. Do you have any ETA? 1 month? 6 months ? 1 year? No, but definitely closer to 1 year than either of the other options at this point. -Greg -- To unsubscribe from this list: send the line unsubscribe ceph-devel in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html