[ https://issues.apache.org/jira/browse/HADOOP-6904?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12969155#action_12969155 ]
Sanjay Radia commented on HADOOP-6904: -------------------------------------- The client side does not know what is *newly* added or deleted. All it has is client and server side list of methods. The client also knows that the major number has not changed (ie getProxy succeeded)-- ie the calls that work should be called. * For the optimized methods, an upper layer (say Hdfs or DfsClient) will check to see if optimal method is available and if not it will call the non-optimized method; for this one has to consult the list of server methods. This matches the case with the previous M-m# approach where one compared the minor number to see if it supports the optimized method (see my list example above). * For the rest of the methods , simply make the call; consulting the list of server methods is an optimization for the failure case. Recall that for the previous M-m# approach we did NOT compare the m# but simply make the call. (Dhruba, did you buy the use of the Major number along with method names?) > A baby step towards inter-version communications between dfs client and > NameNode > -------------------------------------------------------------------------------- > > Key: HADOOP-6904 > URL: https://issues.apache.org/jira/browse/HADOOP-6904 > Project: Hadoop Common > Issue Type: New Feature > Components: ipc > Affects Versions: 0.22.0 > Reporter: Hairong Kuang > Assignee: Hairong Kuang > Fix For: 0.22.0 > > Attachments: majorMinorVersion.patch, majorMinorVersion1.patch, > rpcVersion.patch, rpcVersion1.patch > > > Currently RPC communications in Hadoop is very strict. If a client has a > different version from that of the server, a VersionMismatched exception is > thrown and the client can not connect to the server. This force us to update > both client and server all at once if a RPC protocol is changed. But sometime > different versions do not mean the client & server are not compatible. It > would be nice if we could relax this restriction and allows us to support > inter-version communications. > My idea is that DfsClient catches VersionMismatched exception when it > connects to NameNode. It then checks if the client & the server is > compatible. If yes, it sets the NameNode version in the dfs client and allows > the client to continue talking to NameNode. Otherwise, rethrow the > VersionMismatch exception. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.