[
https://issues.apache.org/jira/browse/HADOOP-2188?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12591818#action_12591818
]
rangadi edited comment on HADOOP-2188 at 4/23/08 3:06 PM:
---------------------------------------------------------------
+1. I looked only at the recent change.
We could set {{call.timestamp only inside doRespond() when it is queue to the
responder. This will avoid two calls to getTime() for each RPC in the normal
case... not that it will really be noticeable.
was (Author: rangadi):
+1. I looked only at the recent change.
We could set {{call.timestamp only inside doRespond() when it is queue to the
responder. This will avoid two calls to getTime(), not that it will really be
noticeable..
> RPC should send a ping rather than use client timeouts
> ------------------------------------------------------
>
> Key: HADOOP-2188
> URL: https://issues.apache.org/jira/browse/HADOOP-2188
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs, ipc
> Affects Versions: 0.16.1
> Reporter: Owen O'Malley
> Assignee: Hairong Kuang
> Attachments: ipc-timeout.patch, ipc-timeout1.patch,
> ipc-timeout2.patch, ipc-timeout3.patch, ipc-timeout4.patch,
> ipc-timeout5.patch, ipc-timeout6.patch, rpc-to.patch
>
>
> Current RPC (really IPC) relies on client side timeouts to find "dead"
> sockets. I propose that we have a thread that once a minute (if the
> connection has been idle) writes a "ping" message to the socket. The client
> can detect a dead socket by the resulting error on the write, so no client
> side timeout is required. Also note that the ipc server does not need to
> respond to the ping, just discard it.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
