[
https://issues.apache.org/jira/browse/HDFS-9228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Bob Hansen updated HDFS-9228:
-----------------------------
Attachment: HDFS-9228.HDFS-8707.004.patch
[~James Clampffer]: Thanks for the feedback.
bq. It looks like RpcConnection::shared_from_this is called frequently but I
wasn't able to find a make_shared<RpcConnection> though maybe I just missed it.
It wass passed into shared_ptr.reset each time, but I changed the interface to
enforce that. Good catch.
bq. In rpc_connection.h there are several instances of a lambda capturing
'this' and a shared_ptr to 'this'. Is the shared_ptr there to keep things alive
and this captured to call methods without using shared_this->method?
It's a good question. I didn't have a good place to document it.
The shared_this is to maintain the lifecycle, and the independent this is to
establish access. That code was failing because we were trying to call
protected/private members from an external lambda; pushing in this allows
private access.
bq. Factoring Request out of RpcConnection looks good but I'd like to have a
short comments for each class that says what their responsibilities are.
Added comments along with threading models and code to enforce the threading
models. The code is a safer place for it.
bq. Perhaps change Callback as typedefed in rpc_engine.h to "RpcCallback" or
similar.
Good call. Fixed.
[~wheat9]: I added an abstract base class and an implementation that
encapsulates the policies (no retry and fixed dealy for now). It's a clone of
the Java interface, so I assume it will have the expressiveness that we need.
Please let me know what you think.
> libhdfs++ should respect NN retry configuration settings
> --------------------------------------------------------
>
> Key: HDFS-9228
> URL: https://issues.apache.org/jira/browse/HDFS-9228
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: hdfs-client
> Reporter: Bob Hansen
> Assignee: Bob Hansen
> Attachments: HDFS-9228.HDFS-8707.001.patch,
> HDFS-9228.HDFS-8707.002.patch, HDFS-9228.HDFS-8707.003.patch,
> HDFS-9228.HDFS-8707.004.patch
>
>
> Handle the use case of temporary network or NN hiccups and have a
> configurable number of retries for NN operations.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
