Hi Vitaly, I've never replicated that many buckets, but pretty sure you can override the bucket limit thru REST API calls. You could test this out, for instance on a cloud service, AWS, GCE pretty inexpensively.
-B On Monday, May 11, 2015 at 9:17:58 AM UTC-4, Vitaly Elyashev wrote: > > Thanks Brian for your help. > My problem is that from security purpose I cannot replicate data that not > site related. > But even if I do so, still I need hundred sites replicate to my site. Are > you tried to replicate number of buckets to same bucket? > Because if its not possible - I still need hundred buckets. > > On Friday, May 8, 2015 at 7:57:10 PM UTC+3, Brian Jones wrote: >> >> I've used XDCR quite a bit. The latest Community version, XDCR seems much >> more stable. We seem to have XDCR dialed in for our data. We move many >> hundreds of MB daily across a few data centers. I remember previous version >> of CB would allow you to override the 10 bucket limit, with the side effect >> of a performance drop. I would suggest creating ONE bucket that is XDCR and >> create unique keys per web site. Saves your bucket capacity for other >> things and won't impact performance like trying to move 100 buckets. >> >> >> On Monday, May 4, 2015 at 10:07:15 AM UTC-4, Vitaly Elyashev wrote: >>> >>> Hi, >>> We're considering to use Couchbase as replicated cache across data >>> centers (using XDCR). >>> >>> We don't have huge amounts of data and we not going change it too much, >>> but we want to support a lot of data centers - it can be around hundred. >>> So single site should be available to replicate to around 100 hundred of >>> different sites and receive updates from these sites as well. >>> There are only few sites that need to replicate to all others - most of >>> them will replicate to only couple of sites. >>> >>> We intended to create bucket per site to be replicated and another >>> bucket to have site updates. So for 100 sites we need 200 buckets, each of >>> them or replicating to or replicating from. >>> >>> In Couchbase docs we found that there are some issues with instance, >>> having more than 10 buckets. Again we don't need greatest performance >>> during the replication, but we do want to use couchbase as local db and >>> cache and performing well during queries. >>> >>> And the question: is Couchbase fit our needs and what best configuration? >>> >>> Best, >>> Vitaly >>> >> -- You received this message because you are subscribed to the Google Groups "Couchbase" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
