Author: jstephens
Date: 2010-08-12 09:40:49 +0200 (Thu, 12 Aug 2010)
New Revision: 22202
Log:
Changelog entries for Cluster bugfixes:
BUG#53987, BUG#55715,
BUG#55839, BUG#55906
Modified:
trunk/dynamic-docs/changelog/mysqld-2.xml
Modified: trunk/dynamic-docs/changelog/mysqld-2.xml
===================================================================
--- trunk/dynamic-docs/changelog/mysqld-2.xml 2010-08-11 22:00:37 UTC (rev 22201)
+++ trunk/dynamic-docs/changelog/mysqld-2.xml 2010-08-12 07:40:49 UTC (rev 22202)
Changed blocks: 1, Lines Added: 140, Lines Deleted: 0; 3753 bytes
@@ -9,6 +9,146 @@
<logentry entrytype="bug">
<tags>
+ <highlight type="clusterapi"/>
+ <highlight type="importantchange"/>
+ <manual type="EINTR"/>
+ <manual type="sockets"/>
+ </tags>
+
+ <bugs>
+ <fixes bugid="55906"/>
+ </bugs>
+
+ <versions>
+ <version ver="5.1.47-ndb-6.3.36"/>
+ <version ver="5.1.47-ndb-7.0.17"/>
+ <version ver="5.1.47-ndb-7.1.6"/>
+ </versions>
+
+ <message>
+
+ <para>
+ The poll and select calls made by the MGM API were not
+ interrupt-safe; that is, a signal caught by the process while
+ waiting for an event on one or more sockets returned error -1
+ with <literal>errno</literal> set to
+ <errorcode>EINTR</errorcode>. This caused problems with MGM API
+ functions such as <literal>ndb_mgm_logevent_get_next()</literal>
+ and <literal>ndb_mgm_get_status2()</literal>.
+ </para>
+
+ <para>
+ To fix this problem, the internal
+ <literal>ndb_socket_poller::poll()</literal> function has been
+ made <errorcode>EINTR</errorcode>-safe.
+ </para>
+
+ <para>
+ The old version of this function has been retained as
+ <literal>poll_unsafe()</literal>, for use by those parts of NDB
+ that do not need the <errorcode>EINTR</errorcode>-safe version
+ of the function.
+ </para>
+
+ </message>
+
+ </logentry>
+
+ <logentry entrytype="bug">
+
+ <tags>
+ <highlight type="cluster"/>
+ <manual type="DBTC"/>
+ <manual type="DBDIH"/>
+ </tags>
+
+ <bugs>
+ <fixes bugid="55715"/>
+ </bugs>
+
+ <versions>
+ <version ver="5.1.47-ndb-6.3.36"/>
+ <version ver="5.1.47-ndb-7.0.17"/>
+ <version ver="5.1.47-ndb-7.1.6"/>
+ </versions>
+
+ <message>
+
+ <para>
+ When another data node failed, a given data node
+ <literal>DBTC</literal> kernel block could time out while
+ waiting for <literal>DBDIH</literal> to signal commits of
+ pending transactions, leading to a crash. Now in such cases the
+ timeout generates a prinout, and the data node continues to
+ operate.
+ </para>
+
+ </message>
+
+ </logentry>
+
+ <logentry entrytype="bug">
+
+ <tags>
+ <highlight type="cluster"/>
+ <manual type="timeouts"/>
+ <manual type="logs"/>
+ </tags>
+
+ <bugs>
+ <fixes bugid="53987"/>
+ </bugs>
+
+ <versions>
+ <version ver="5.1.47-ndb-6.3.36"/>
+ <version ver="5.1.47-ndb-7.0.17"/>
+ <version ver="5.1.47-ndb-7.1.6"/>
+ </versions>
+
+ <message>
+
+ <para>
+ An excessive number of timeout warnings (normally used only for
+ debugging) were written to the data node logs.
+ </para>
+
+ </message>
+
+ </logentry>
+
+ <logentry entrytype="bug">
+
+ <tags>
+ <highlight type="cluster"/>
+ <manual type="ndb_config"/>
+ <manual type="--configinfo option"/>
+ </tags>
+
+ <bugs>
+ <fixes bugid="55839"/>
+ </bugs>
+
+ <versions>
+ <version ver="5.1.47-ndb-7.0.17"/>
+ <version ver="5.1.47-ndb-7.1.6"/>
+ </versions>
+
+ <message>
+
+ <para>
+ The TCP configuration parameters <literal>HostName1</literal>
+ and <literal>HostName2</literal> were not displayed in the
+ output of <command>ndb_config</command>
+ <option role="ndb_config">--configinfo</option>.
+ </para>
+
+ </message>
+
+ </logentry>
+
+ <logentry entrytype="bug">
+
+ <tags>
<manual type="Build"/>
<manual type="Solaris"/>
</tags>
| Thread |
|---|
| • svn commit - mysqldoc@docsrva: r22202 - trunk/dynamic-docs/changelog | jon.stephens | 12 Aug |
