Apache
/
hbase
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Edit of what is done on WAL exceptions 

git-svn-id: https://svn.apache.org/repos/asf/hbase/trunk@992324 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Michael Stack 2010-09-03 15:20:19 +00:00
parent 39e213e62d
commit 33257e051e
1 changed files with 59 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

81
src/docbkx/book.xml
View File

@ -7,7 +7,20 @@
xmlns:html="http://www.w3.org/1999/xhtml"
xmlns:db="http://docbook.org/ns/docbook">
<info>
<title>HBase Book<?eval ${project.version}?></title>
<title>The <link xlink:href="http://www.hbase.org">HBase</link>
Book</title>
<revhistory>
<revision>
<date />
<revdescription>Initial layout</revdescription>
<revnumber>
<?eval ${project.version}?>
</revnumber>
</revision>
</revhistory>
</info>
<chapter xml:id="getting_started">
@ -595,34 +608,58 @@
<title>The WAL</title>
<subtitle>HBase's<link
xlink:href="http://en.wikipedia.org/wiki/Write-ahead_logging"> <link
linkend="???">Write-Ahead Log</link></link></subtitle>
xlink:href="http://en.wikipedia.org/wiki/Write-ahead_logging"> Write-Ahead
Log</link></subtitle>
<para>Each RegionServer adds updates to its <link linkend="???">WAL</link>
first, and then to memory.</para>
<para></para>
<section>
<title>What is the purpose of the HBase WAL</title>
<para>The HBase WAL is...</para>
</section>
<section>
<title>How EOFExceptions are treated when splitting a crashed
RegionServers' WALs </title>
<title>WAL splitting</title>
<para>If we get an EOF while splitting logs, we proceed with the split
even when <varname>hbase.hlog.split.skip.errors</varname> ==
<constant>false</constant>. An EOF while reading the last log in the set
of files to split is near-guaranteed since the RegionServer likely
crashed mid-write of a record. But we'll continue even if we got an EOF
reading other than the last file in the set.<footnote>
<para>For background, see <link
xlink:href="https://issues.apache.org/jira/browse/HBASE-2643">HBASE-2643
Figure how to deal with eof splitting logs</link></para>
</footnote></para>
<subtitle>How edits are recovered from a crashed RegionServer</subtitle>
<para>When a RegionServer crashes, it will lose its ephemeral lease in
ZooKeeper...TODO</para>
<section>
<title><varname>hbase.hlog.split.skip.errors</varname></title>
<para>When set to <constant>true</constant>, the default, any error
encountered splitting will be logged, the problematic WAL will be
moved into the <filename>.corrupt</filename> directory under the hbase
<varname>rootdir</varname>, and processing will continue. If set to
<constant>false</constant>, the exception will be propagated and the
split logged as failed.<footnote>
<para>See <link
xlink:href="https://issues.apache.org/jira/browse/HBASE-2958">HBASE-2958
When hbase.hlog.split.skip.errors is set to false, we fail the
split but thats it</link>. We need to do more than just fail split
if this flag is set.</para>
</footnote></para>
</section>
<section>
<title>How EOFExceptions are treated when splitting a crashed
RegionServers' WALs</title>
<para>If we get an EOF while splitting logs, we proceed with the split
even when <varname>hbase.hlog.split.skip.errors</varname> ==
<constant>false</constant>. An EOF while reading the last log in the
set of files to split is near-guaranteed since the RegionServer likely
crashed mid-write of a record. But we'll continue even if we got an
EOF reading other than the last file in the set.<footnote>
<para>For background, see <link
xlink:href="https://issues.apache.org/jira/browse/HBASE-2643">HBASE-2643
Figure how to deal with eof splitting logs</link></para>
</footnote></para>
</section>
</section>
</chapter>
<appendix>
<title></title>
<para></para>
</appendix>
</book>