Thanks for kind reply! Have a nice day!
:) 2017-04-11 8:47 GMT+09:00 Dan Burkert <[email protected]>: > Oops, the tablet ID I used in the example is ' > 4398cf80d68141cdbdae882e97b6da45', not 'c5299ec14315401a89316b62afad5877'. > > - Dan > > On Mon, Apr 10, 2017 at 4:34 PM, Dan Burkert <[email protected]> > wrote: > >> Kudu does not yet have a way to request tablet rebalancing, but we do >> have a few tools for balancing tablets manually. >> >> For example, if you had a tablet 'c5299ec14315401a89316b62afad5877' >> which you wanted to remove from an old tserver >> 'c5299ec14315401a89316b62afad5877' and add to a new tserver >> '4e6cdb2609fd488586fb526b0646db03', you could perform the following >> steps using the kudu CLI tool: >> >> kudu tablet change-config add-replica <master-addrs> >> 4398cf80d68141cdbdae882e97b6da45 4e6cdb2609fd488586fb526b0646db03 VOTER >> kudu tablet change-config drop-replica <master-addrs> >> 4398cf80d68141cdbdae882e97b6da45 c5299ec14315401a89316b62afad5877 >> >> This will create a new replica on the new server, then drop the old >> replica on the old server. In-between adding the new replica and dropping >> the old replica, you should double check that the tablet is healthy with >> the ksck tool, e.g.: >> >> kudu cluster ksck --tablets 4398cf80d68141cdbdae882e97b6da45 >> <master-addr> >> >> This is a little tedious since you have to choose locations yourself, so >> hopefully Kudu will be getting a more automated tool in the future to make >> this seamless. >> >> - Dan >> >> On Mon, Apr 10, 2017 at 4:15 AM, 기준 <[email protected]> wrote: >> >>> Hi! >>> >>> My team using 20 nodes for kudu cluster. >>> (I'm using Apache Kudu 1.2 on CDH 5.10.) >>> >>> As data grows up, our team planned to add new nodes to existing cluster. >>> >>> Unfortunately, it's seems there are no tools for move tablets to another >>> or redistributes to new nodes. >>> >>> For now i'm thinking the only way to redistribute all old tables is to >>> shutdown each old one >>> and wait until old tablet goes to new one. (I think this is not >>> efficient, not safe) >>> >>> Is there any recommended scale out stretegy for now? >>> And i'm wondering is there any plan to add tools for these? (like >>> kafka-reassign-partitions.sh) >>> >>> Thanks!! >>> >>> Have a nice day! >>> >> >> >
