Hi Leng, Can you share your distributed cfg and the exact line of code you use to do the save?
Lvc@ On 28 January 2015 at 03:20, Leng Sheng Hong <[email protected]> wrote: > Yes, but this seems to be the same as 1.7? I was wondering now what does > auto sharding means in 2.0, so inserting to Class MyTable will not > automatically goes to the nodes connected in round robin fashion? but > always goes to the default node? For me, it always goes to the other node > rather than the default node. So I am a bit confused with how distributed > sharding works when inserting to a Class without specifying a cluster. > > On Wednesday, 28 January 2015 06:13:12 UTC+8, Lvc@ wrote: >> >> Have you read http://www.orientechnologies.com/docs/last/orientdb.wiki/ >> Distributed-Sharding.html ? >> >> On 27 January 2015 at 05:53, Leng Sheng Hong <[email protected]> wrote: >> >>> I have a question on how auto sharding work. This is my assumption on >>> v2.0 using the default configurations: >>> >>> 1) setup database on node 1, start in distributed mode >>> 2) start node 2 in distributed mode >>> 3) node 1 will send and deploy DB on node 2 >>> 4) app insert a new document of a class into the DB will automatically >>> go into node 1, then node 2 in round robin fashion. >>> >>> when i look at the databases/mydb/distributed-config.json it contains >>> new clusters of node 2 which is handled on both node?? >>> { >>> "@type": "d", >>> "@version": 0, >>> "version": 14, >>> "autoDeploy": true, >>> "hotAlignment": false, >>> "executionMode": "undefined", >>> "readQuorum": 1, >>> "writeQuorum": 2, >>> "failureAvailableNodesLessQuorum": false, >>> "readYourWrites": true, >>> "clusters": { >>> "@type": "d", >>> "@version": 0, >>> "internal": { >>> "@type": "d", >>> "@version": 0 >>> }, >>> "index": { >>> "@type": "d", >>> "@version": 0 >>> }, >>> "*": { >>> "@type": "d", >>> "@version": 0, >>> "servers": [ >>> "node1", >>> "node2", >>> "<NEW_NODE>" >>> ] >>> }, >>> "orole_node2": { >>> "@type": "d", >>> "@version": 0, >>> "servers": [ >>> "node1", >>> "node2", >>> "<NEW_NODE>" >>> ] >>> }, >>> "mytable_node2": { >>> "@type": "d", >>> "@version": 0, >>> "servers": [ >>> "node1", >>> "node2", >>> "<NEW_NODE>" >>> ] >>> }, >>> >>> >>> Now if i tried to insert 10 records from my app with the Java api >>> document.save() into mytable >>> all the records i get is saved with cluster ID of node2. >>> >>> So, how is this data distributed? It never uses node1 at all. >>> >>> >>> >>> -- >>> >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "OrientDB" 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. >>> >> >> -- > > --- > You received this message because you are subscribed to the Google Groups > "OrientDB" 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. > -- --- You received this message because you are subscribed to the Google Groups "OrientDB" 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.
