Github user joshelser commented on a diff in the pull request:

    https://github.com/apache/incubator-ratis/pull/4#discussion_r212007378
  
    --- Diff: 
ratis-logservice/src/main/java/org/apache/ratis/logservice/api/ArchivedLog.java 
---
    @@ -0,0 +1,43 @@
    +/**
    + * Licensed to the Apache Software Foundation (ASF) under one
    + * or more contributor license agreements.  See the NOTICE file
    + * distributed with this work for additional information
    + * regarding copyright ownership.  The ASF licenses this file
    + * to you under the Apache License, Version 2.0 (the
    + * "License"); you may not use this file except in compliance
    + * with the License.  You may obtain a copy of the License at
    + *
    + *     http://www.apache.org/licenses/LICENSE-2.0
    + *
    + * Unless required by applicable law or agreed to in writing, software
    + * distributed under the License is distributed on an "AS IS" BASIS,
    + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
    + * See the License for the specific language governing permissions and
    + * limitations under the License.
    + */
    +package org.apache.ratis.logservice.api;
    +
    +/**
    + * A {@link LogStream} which has been archived in some external
    + * system. This interface is parameterized to allow for implementations
    + * to use their own class to encapsulate how to find the archived log.
    + *
    + * In the majority of cases, this should be transparent to end-users, as
    + * the {@link LogStream} should hide the fact that this even exists.
    + * TODO maybe that means this should be client-facing at all?
    + *
    + * @param <T> A referent to the log on the external system.
    + */
    +public interface ArchivedLog<T> extends AutoCloseable {
    --- End diff --
    
    Yeah, I was mulling over whether or not to even include it in this initial 
drop.
    
    I like the idea of a consistent API to read from Log, regardless of it 
being "active" (still in the state machine) or "archived" (exported to some 
remote storage).
    
    > Log archiving is the feature HBase requires
    
    I think that this is a generally reusable feature that users would like. It 
would help keep the state machine "small" and consistent performance.
    
    > HBase client will create archive subdirectory and will be moving old 
files to this subdirectory
    
    I think automated archival should be a feature that LogService provides. 
HBase should be unaware that we are moving "segments" of data around.
    
    > I want to enhance LogService with FileSystem-like API:
    
    I have changes that have been sitting locally for too long. I'm 
prioritizing getting them finished and then update this. I think being able to 
think of the LogService like a FileSystem is nice. I'm curious if you have put 
more thought about how the notion of "directories" might be used :)


---

Reply via email to