[ https://issues.apache.org/jira/browse/HDFS-7858?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun Suresh updated HDFS-7858: ------------------------------ Attachment: HDFS-7858.7.patch Thanks [~arpitagarwal] and [~jingzhao] for your reviews. Uploading patch addressing your suggestions. w.r.t. Using CompletionService. Yup.. thanks, it did make the implementation more readable. bq. I didn't understand the call to super.performFailover in RequestHedgingProxyProvider#getProxy. Yeah.. i wanted to increment the proxy index. Agreed, it does look out of place. Ive created an explicit method to make it more readable. bq. For RequestHedgingProxyProvider#performFailover, if the original successfulProxy is not null, we can exclude it for the next time retry. So, in the case of the ReqHedgingProxy, {{performFailover}} will be called only if ALL the proxies have failed (with retry/failover_and_retry.. ), in which case, in the next attempt, the request will be again sent to all the namenodes, so dont think it makes sense to exclude it. bq. new LinkedList<RetryAction> - explicit type argument redundant. Oh.. I was thinking we should keep trunk Java 7 compilable ? > Improve HA Namenode Failover detection on the client > ---------------------------------------------------- > > Key: HDFS-7858 > URL: https://issues.apache.org/jira/browse/HDFS-7858 > Project: Hadoop HDFS > Issue Type: Improvement > Reporter: Arun Suresh > Assignee: Arun Suresh > Labels: BB2015-05-TBR > Attachments: HDFS-7858.1.patch, HDFS-7858.2.patch, HDFS-7858.2.patch, > HDFS-7858.3.patch, HDFS-7858.4.patch, HDFS-7858.5.patch, HDFS-7858.6.patch, > HDFS-7858.7.patch > > > In an HA deployment, Clients are configured with the hostnames of both the > Active and Standby Namenodes.Clients will first try one of the NNs > (non-deterministically) and if its a standby NN, then it will respond to the > client to retry the request on the other Namenode. > If the client happens to talks to the Standby first, and the standby is > undergoing some GC / is busy, then those clients might not get a response > soon enough to try the other NN. > Proposed Approach to solve this : > 1) Since Zookeeper is already used as the failover controller, the clients > could talk to ZK and find out which is the active namenode before contacting > it. > 2) Long-lived DFSClients would have a ZK watch configured which fires when > there is a failover so they do not have to query ZK everytime to find out the > active NN > 2) Clients can also cache the last active NN in the user's home directory > (~/.lastNN) so that short-lived clients can try that Namenode first before > querying ZK -- This message was sent by Atlassian JIRA (v6.3.4#6332)