Author: jstephens
Date: 2007-03-22 02:59:06 +0100 (Thu, 22 Mar 2007)
New Revision: 5496
Log:
Undo logs no longer used in 5.1 (except for DD storage) (Jonas & Roland)
Modified:
trunk/refman-5.1/mysql-cluster.xml
Modified: trunk/refman-5.1/mysql-cluster.xml
===================================================================
--- trunk/refman-5.1/mysql-cluster.xml 2007-03-21 18:20:58 UTC (rev 5495)
+++ trunk/refman-5.1/mysql-cluster.xml 2007-03-22 01:59:06 UTC (rev 5496)
Changed blocks: 11, Lines Added: 41, Lines Deleted: 30; 6877 bytes
@@ -3749,9 +3749,9 @@
<para>
This parameter specifies the directory where all files
- created for metadata, REDO logs, UNDO logs and data files
- are placed. The default is the directory specified by
- <literal>DataDir</literal>.
+ created for metadata, REDO logs, UNDO logs (for Disk Data
+ tables) and data files are placed. The default is the
+ directory specified by <literal>DataDir</literal>.
<emphasis role="bold">Note</emphasis>: This directory must
exist before the <command>ndbd</command> process is
initiated.
@@ -5521,13 +5521,10 @@
<literal>NoOfDiskPagesToDiskAfterRestartTUP</literal> to a
value of <literal>20</literal> entails writing 1.6MB in
data pages to disk each second during a local checkpoint.
- This value includes the writing of UNDO log records for
- data pages. That is, this parameter handles the limitation
- of writes from data memory. UNDO log records for index
- pages are handled by the parameter
- <literal>NoOfDiskPagesToDiskAfterRestartACC</literal>.
- (See the entry for <literal>IndexMemory</literal> for
- information about index pages.)
+ That is, this parameter handles the limitation of writes
+ from data memory. (See the entry for
+ <literal>IndexMemory</literal> for information about index
+ pages.)
</para>
<para>
@@ -5707,9 +5704,9 @@
All update activities need to be logged. The REDO log makes it
possible to replay these updates whenever the system is
restarted. The NDB recovery algorithm uses a
- <quote>fuzzy</quote> checkpoint of the data together with the
- UNDO log, and then applies the REDO log to play back all
- changes up to the restoration point.
+ <quote>fuzzy</quote> checkpoint of the data, and then applies
+ the REDO log to play back all changes up to the restoration
+ point.
</para>
<para>
@@ -5746,8 +5743,9 @@
<emphasis role="bold">Note</emphasis>: The
<literal>UndoIndexBuffer</literal> and
<literal>UndoDataBuffer</literal> parameters which appeared in
- earlier versions of MySQL Cluster are no longer necessary in
- MySQL 5.1.
+ earlier versions of MySQL Cluster are not necessary in MySQL
+ 5.1, since undo logs are no longer kept on disk for in-memory
+ tables.
</para>
<para id="mysql-cluster-controlling-log-messages">
@@ -8356,12 +8354,12 @@
<para>
It is also important to remember that the disk also handles
- writes to the REDO log and UNDO log, so if you find that the
- amount of data being written to disk as detemined by the values
- of <literal>NoOfDiskPagesToDiskAfterRestartACC</literal> and
+ writes to the REDO log, so if you find that the amount of data
+ being written to disk as detemined by the values of
+ <literal>NoOfDiskPagesToDiskAfterRestartACC</literal> and
<literal>NoOfDiskPagesToDiskAfterRestartTUP</literal> is
approaching the amount of disk bandwidth available, you may wish
- to increase the time between local checkpoints.
+ to increase the time between local checkpoints.
</para>
<para>
@@ -8433,14 +8431,6 @@
than ample in this case.
</para>
- <para>
- A copy of each altered table record is kept in the UNDO log. In
- the scenario discussed above, the UNDO log would not require any
- more space than what is provided by the default seetings.
- However, given the size of disks, it is sensible to allocate at
- least 1 GB for it.
- </para>
-
</section>
</section>
@@ -10520,7 +10510,7 @@
<listitem>
<para>
- Read data from the local checkpoint and undo logs.
+ Read data from the local checkpoint.
</para>
</listitem>
@@ -11658,12 +11648,18 @@
<entry><literal>INFO</literal></entry>
<entry>Start of LCP: data written to disk</entry>
</row>
+<!--
<row>
+ <remark role="note">
+ [js] No longer applies since there are no more undo
+ logs in 5.1, right?
+ </remark>
<entry>Report undo log blocked</entry>
<entry>7</entry>
<entry><literal>INFO</literal></entry>
<entry>UNDO logging blocked; buffer near overflow</entry>
</row>
+-->
</tbody>
</tgroup>
</informaltable>
@@ -11709,12 +11705,18 @@
<entry><literal>INFO</literal></entry>
<entry>Blocks received after completion of restart</entry>
</row>
+<!--
+ <remark role="note">
+ [js] No longer applies since 5.1 doesn't have undo logs,
+ right?
+ </remark>
<row>
<entry>Undo records executed</entry>
<entry>15</entry>
<entry><literal>INFO</literal></entry>
<entry/>
</row>
+-->
<row>
<entry>New REDO log started</entry>
<entry>10</entry>
@@ -11734,8 +11736,9 @@
<entry>Node has been refused for inclusion in the cluster</entry>
<entry>8</entry>
<entry><literal>INFO</literal></entry>
- <entry>Node cannot be included in cluster due to misconfiguration, inability to
- establish communication, or other problem</entry>
+ <entry>Node cannot be included in cluster due to
+ misconfiguration, inability to establish
+ communication, or some other problem</entry>
</row>
<row>
<entry>data node neighbors</entry>
@@ -18463,6 +18466,14 @@
one or more undo log files to it (an undo log file is also
referred as an <firstterm>undofile</firstterm>).
</para>
+
+ <note>
+ <para>
+ In MySQL 5.1, undo log files are necessary only for Disk
+ Data tables. They are no longer used for tables that are
+ stored in memory.
+ </para>
+ </note>
</listitem>
<listitem>
| Thread |
|---|
| • svn commit - mysqldoc@docsrva: r5496 - trunk/refman-5.1 | jon | 22 Mar |
