[
https://issues.apache.org/jira/browse/HBASE-24471?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang updated HBASE-24471:
------------------------------
Hadoop Flags: Incompatible change,Reviewed (was: Incompatible change)
Release Note:
Move all the meta initialization code in MasterFileSystem and HRegionServer to
InitMetaProcedure. Add a new step for InitMetaProcedure called
INIT_META_WRITE_FS_LAYOUT to place the moved code.
This is an incompatible change, but should not have much impact.
InitMetaProcedure will only be executed once when bootstraping a fresh new
cluster, so typically this will not effect rolling upgrading. And even if you
hit this problem, as long as InitMetaProcedure has not been finished, we can
make sure that there is no user data in the cluster, you can just clean up the
cluster and try again. There will be no data loss.
> The way we bootstrap meta table is confusing
> --------------------------------------------
>
> Key: HBASE-24471
> URL: https://issues.apache.org/jira/browse/HBASE-24471
> Project: HBase
> Issue Type: Improvement
> Components: master, meta, proc-v2
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Priority: Major
> Fix For: 3.0.0-alpha-1, 2.4.0
>
>
> I used to thought the work is only done in the InitMetaProcedure but in fact
> we have several pieces of code related to this which are placed in different
> classes.
> In HRegionServer.initializeFileSystem, we will test whether we are HMaster.
> If so, we will call FSTableDescriptors.tryUpdateMetaTableDescriptor. And in
> MasterFileSystem.checkRootDir, we will check if meta region is available. If
> not, we will bootstrap it, and it will also call
> FSTableDescriptors.tryUpdateMetaTableDescriptor.
> The problem here is that, HRegionServer.initializeFileSystem is called in the
> constructor of HRegionServer, which means it will be executed before
> MasterFileSystem.checkRootDir, as we will only create the MasterFileSystem
> for active master...
> Ideally I prefer we move all the bootstrap code into InitMetaProcedure, which
> will be much cleaner. If this can not be done, at least we could move the
> code into MasterFileSystem.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
