| George Herson 2005-09-29, 8:23 pm |
| -------------- 08010900030902020007
0200
Content-Type: multipart/alternative;
boundary="------------ 01000807010301020203
0808"
-------------- 01000807010301020203
0808
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
James G. Sack (jim) wrote:
>On Mon, 2005-09-19 at 13:14 -0400, George Herson wrote:
>
>
>
>George, LVM snapshots are generally intended to be short-lived --
>
>
<snip>
>..jim
>
>
>
Jim,
I didn't word my question quite right because I was only guessing at
what a LVM snapshot was. Moreover, what you're saying is all correct.
However, I was not suggesting that the snapshot be kept around once the
backup is made.
Let's go to article "What is a Logical Volume Manager (LVM) snapshot and
how do I use it?
<http://kbase.redhat.com/faq/dml_fet...ame=AS/ES/WS%20
Basics&Id=&Instance=&Shared=>"
in the RedHat k'base
<http://www.redhat.com/apps/support/knowledgebase/>. It says "After
performing the backup of the snapshot partition we release the
snapshot". This implies, at least to me, that one doesn't need a 2nd
database server or to do a mysqldump (your steps 5-7). Instead, we just
tar cv /mnt/ops/dbbackup (to use the article's example name for the
mounted snapshot), save the tape, and dispense with the snapshot.
Wouldn't that work? MySQL keeps its data in files already, so why is it
necessary to mysqldump it? Are you only trying to avoid having to also
backup the mysqld version that wrote the data files to ensure that these
can be read later?
George
-------------- 01000807010301020203
0808
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
James G. Sack (jim) wrote:<br>
<blockquote cite="mid1127156031.4962.43.camel@jgs4.ino.pvt" type="cite">
<pre wrap="">On Mon, 2005-09-19 at 13:14 -0400, George Herson wrote:
</pre>
<blockquote type="cite">
<pre wrap="">James G. Sack (jim) wrote:
</pre>
<blockquote type="cite">
<pre wrap="">On Mon, 2005-09-19 at 12:15 -0400, George Herson wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Dear Jim,
Re: your post at <a class="moz-txt-link-freetext" href="http://lists.mysql.com/mysql/189058">http://lists.mysql.com/mysql/189058</a>, why bother
creating the mysqldump if you already have the snapshot? Why not just
backup the snapshot?
</pre>
</blockquote>
<pre wrap="">[...]
The output of mysqldump *is* the backup. If the db goes away, it can be
restored with (something like) "mysql <dumpfile.2005-09-01".
The dump operation is run periodically, and some number of back versions
can be kept around (or offloaded) for archival value.
</pre>
</blockquote>
<pre wrap="">Yes, but can't you also save your snapshot instead, then, when/if you
want, restore it,
"4. mount the snapshot
5. load a second database server daemon accessing the db within the
snapshot (with a suitable alternate my.cnf file)
6. perform mysqldump operation on the snapshot-db" ??
...
</pre>
</blockquote>
<pre wrap=""><!---->
George, LVM snapshots are generally intended to be short-lived --
</pre>
</blockquote>
<snip><br>
<blockquote cite="mid1127156031.4962.43.camel@jgs4.ino.pvt" type="cite">
<pre wrap="">..jim
</pre>
</blockquote>
Jim,<br>
<br>
I didn't word my question quite right because I was only guessing at
what a LVM snapshot was. Moreover, what you're saying is all correct.
However, I was not suggesting that the snapshot be kept around once the
backup is made. <br>
<br>
Let's go to article "<a
href="http://kbase.redhat.com/faq/dml_fetch.pl?CompanyID=842&ContentID=4097&FaqID=3640& amp;word=What%20is%2
0a%20Logical%20Volum
e%20Manager& amp;faq_template=htt
p://kbase.redhat.com/faq/searchfaq.shtm&topic=80&back_refr=http://kbase.redhat.
com/faq/&topicname=AS/ES/WS%20Basics&Id=&Instance=&Shared=">What
is a Logical Volume Manager (LVM) snapshot and how do I use it?</a>" in
the <a href="http://www.redhat.com/apps/support/knowledgebase/">RedHat
k'base</a>. It says "After performing the backup of the snapshot
partition we release the snapshot". This implies, at least to me, that
one doesn't need a 2nd database server or to do a mysqldump (your steps
5-7). Instead, we just <tt>tar cv /</tt><span class="command"><tt>mnt/ops/dbbackup</tt>
(to use the article's example name for the mounted snapshot), save the
tape, and dispense with the snapshot. Wouldn't that work?</span> MySQL
keeps its data in files already, so why is it necessary to mysqldump
it? Are you only trying to avoid having to also backup the mysqld
version that wrote the data files to ensure that these can be read
later?<tt></tt><br>
<br>
George<br>
</body>
</html>
-------------- 01000807010301020203
0808--
-------------- 08010900030902020007
0200
Content-Type: text/plain; charset=us-ascii
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/mysql? unsub...sie.nctu.edu.tw
-------------- 08010900030902020007
0200--
|