You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hbase.apache.org by Jonathan Gray <jg...@apache.org> on 2010/11/29 20:04:33 UTC
Review Request: Need to retain DeadServers to ensure we don't allow
previously expired RS instances to rejoin cluster
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://review.cloudera.org/r/1259/
-----------------------------------------------------------
Review request for hbase and stack.
Summary
-------
We currently let go of dead servers once we finish their shutdown. We should hang on to them longer to deal with things like network partitions.
I'm not a fan of SoftReferences so I decided on another approach. DeadServers now has a maximum number of servers to hold on to in the set (default 100). Once it reaches the max, it evicts the oldest.
More code than I had hoped but nothing too crazy.
This addresses bug HBASE-3282.
http://issues.apache.org/jira/browse/HBASE-3282
Diffs
-----
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/DeadServer.java 1040221
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java 1040221
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java 1040221
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/handler/ServerShutdownHandler.java 1040221
Diff: http://review.cloudera.org/r/1259/diff
Testing
-------
Running unit tests now.
Thanks,
Jonathan
Re: Review Request: Need to retain DeadServers to ensure we don't allow
previously expired RS instances to rejoin cluster
Posted by Jonathan Gray <jg...@apache.org>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://review.cloudera.org/r/1259/
-----------------------------------------------------------
(Updated 2010-11-29 11:43:07.682958)
Review request for hbase and stack.
Changes
-------
Makes DeadServers private. It was still accessed by my TestRollingRestart test so I had to make a small change to how that worked.
Also added some additional testing in TestDeadServers that verifies the new boolean check works as expected and also that the max capacity works as expected.
Summary
-------
We currently let go of dead servers once we finish their shutdown. We should hang on to them longer to deal with things like network partitions.
I'm not a fan of SoftReferences so I decided on another approach. DeadServers now has a maximum number of servers to hold on to in the set (default 100). Once it reaches the max, it evicts the oldest.
More code than I had hoped but nothing too crazy.
This addresses bug HBASE-3282.
http://issues.apache.org/jira/browse/HBASE-3282
Diffs (updated)
-----
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/DeadServer.java 1040242
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java 1040242
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java 1040242
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/handler/ServerShutdownHandler.java 1040242
branches/0.90/src/test/java/org/apache/hadoop/hbase/master/TestDeadServer.java 1040242
branches/0.90/src/test/java/org/apache/hadoop/hbase/master/TestRollingRestart.java 1040242
Diff: http://review.cloudera.org/r/1259/diff
Testing
-------
Running unit tests now.
Thanks,
Jonathan
Re: Review Request: Need to retain DeadServers to ensure we don't allow
previously expired RS instances to rejoin cluster
Posted by st...@duboce.net.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://review.cloudera.org/r/1259/#review2004
-----------------------------------------------------------
Ship it!
branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
<http://review.cloudera.org/r/1259/#comment6320>
You can make this private now that its no longer referenced by Master?
- stack
On 2010-11-29 11:04:33, Jonathan Gray wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://review.cloudera.org/r/1259/
> -----------------------------------------------------------
>
> (Updated 2010-11-29 11:04:33)
>
>
> Review request for hbase and stack.
>
>
> Summary
> -------
>
> We currently let go of dead servers once we finish their shutdown. We should hang on to them longer to deal with things like network partitions.
>
> I'm not a fan of SoftReferences so I decided on another approach. DeadServers now has a maximum number of servers to hold on to in the set (default 100). Once it reaches the max, it evicts the oldest.
>
> More code than I had hoped but nothing too crazy.
>
>
> This addresses bug HBASE-3282.
> http://issues.apache.org/jira/browse/HBASE-3282
>
>
> Diffs
> -----
>
> branches/0.90/src/main/java/org/apache/hadoop/hbase/master/DeadServer.java 1040221
> branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java 1040221
> branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java 1040221
> branches/0.90/src/main/java/org/apache/hadoop/hbase/master/handler/ServerShutdownHandler.java 1040221
>
> Diff: http://review.cloudera.org/r/1259/diff
>
>
> Testing
> -------
>
> Running unit tests now.
>
>
> Thanks,
>
> Jonathan
>
>