<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Home Fileserver: ZFS snapshots</title>
	<atom:link href="http://breden.org.uk/2008/05/12/home-fileserver-zfs-snapshots/feed/" rel="self" type="application/rss+xml" />
	<link>http://breden.org.uk/2008/05/12/home-fileserver-zfs-snapshots/</link>
	<description>Complexifying simplicity</description>
	<lastBuildDate>Fri, 04 Nov 2011 23:45:48 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.4</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Simon</title>
		<link>http://breden.org.uk/2008/05/12/home-fileserver-zfs-snapshots/comment-page-1/#comment-6117</link>
		<dc:creator>Simon</dc:creator>
		<pubDate>Sat, 11 Apr 2009 10:47:05 +0000</pubDate>
		<guid isPermaLink="false">http://breden.org.uk/2008/05/12/home-fileserver-zfs-snapshots/#comment-6117</guid>
		<description>Hi Haudy,

Thanks for the feedback.

Regarding &#039;ZFS list&#039; reporting (now it&#039;s &#039;zfs list -t snapshot&#039;), I don&#039;t know what the answer is.

I know what you mean about errors often being missed by those that are veterans, and often spotted by newcomers.

Regarding the contradiction, I will elect to keep the text remaining, as it serves to show that unexpected things can happen sometimes. However, to help clarify things, I have added an &#039;update&#039; to explain that in that particular case, I think the reason for the snapshot unexpectedly consuming space was due to me doing an &#039;ls&#039; on the directory, causing the &#039;atime&#039; attribute to get updated, and therefore the space consumed by the snapshot increased. This comment described it well, I think:
http://breden.org.uk/2008/05/12/home-fileserver-backups-from-zfs-snapshots/#comment-1267
where he says:

&lt;blockquote&gt;
The reason why an ls changes the file system is probably because access times or other file / directory attributes are being modified… That would also explain the “growth” in a snapshot in the previous article even though you hadn’t changed the files or added any - by looking at the directories you are affecting them.
&lt;/blockquote&gt;

Thanks also for spotting the error where I stated &#039;5&#039; instead of &#039;4&#039; -- updated now! :)

Cheers,
Simon</description>
		<content:encoded><![CDATA[<p>Hi Haudy,</p>
<p>Thanks for the feedback.</p>
<p>Regarding &#8216;ZFS list&#8217; reporting (now it&#8217;s &#8216;zfs list -t snapshot&#8217;), I don&#8217;t know what the answer is.</p>
<p>I know what you mean about errors often being missed by those that are veterans, and often spotted by newcomers.</p>
<p>Regarding the contradiction, I will elect to keep the text remaining, as it serves to show that unexpected things can happen sometimes. However, to help clarify things, I have added an &#8216;update&#8217; to explain that in that particular case, I think the reason for the snapshot unexpectedly consuming space was due to me doing an &#8216;ls&#8217; on the directory, causing the &#8216;atime&#8217; attribute to get updated, and therefore the space consumed by the snapshot increased. This comment described it well, I think:<br />
<a href="http://breden.org.uk/2008/05/12/home-fileserver-backups-from-zfs-snapshots/#comment-1267" rel="nofollow">http://breden.org.uk/2008/05/12/home-fileserver-backups-from-zfs-snapshots/#comment-1267</a><br />
where he says:</p>
<blockquote><p>
The reason why an ls changes the file system is probably because access times or other file / directory attributes are being modified… That would also explain the “growth” in a snapshot in the previous article even though you hadn’t changed the files or added any &#8211; by looking at the directories you are affecting them.
</p></blockquote>
<p>Thanks also for spotting the error where I stated &#8216;5&#8242; instead of &#8216;4&#8242; &#8212; updated now! <img src='http://breden.org.uk/wordpress/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
<p>Cheers,<br />
Simon</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Haudy Kazemi</title>
		<link>http://breden.org.uk/2008/05/12/home-fileserver-zfs-snapshots/comment-page-1/#comment-5941</link>
		<dc:creator>Haudy Kazemi</dc:creator>
		<pubDate>Sat, 04 Apr 2009 06:26:18 +0000</pubDate>
		<guid isPermaLink="false">http://breden.org.uk/2008/05/12/home-fileserver-zfs-snapshots/#comment-5941</guid>
		<description>Is &#039;ZFS list&#039; still reporting the same way in the current release (assuming you now add the required snapshots parameter)?  Maybe you ran into a minor stats display bug?  This (old) thread might be useful to you: http://www.mail-archive.com/zfs-discuss@opensolaris.org/msg05319.html

Separately, I myself saw a minor bug in the wording of an error message when I was trying out ZFS on NexentaCore 1.0.  I ended up reporting it to the forums, OpenSolaris mailing list, and bugtracker.  In terms of triaging, little stats/text display bugs likely receive lower priority than most other bugs even though they may be low hanging fruit.  I also suspect these bugs are more noticeable to people new to and just learning ZFS rather than ZFS veterans who may not notice little inconsistencies like these.  I&#039;ve run across a parallel situation in some university textbooks where there are internal errors and inconsistencies in example problems that make the problem hard to use or even useless as a learning example.  These issues usually don&#039;t confound experienced people but they sure can throw a curveball for anyone new.

By the way, in your article above, you probably want to re-word this sentence which is self contradictory:
&quot;The snapshot ‘tank/test@3? references only the new modified file ‘b’ and consumes no disk space. Oh, but it seems it does consume some disk space! Why?&quot;

This sentence appears to have a typo where 5 should be replaced with 4:
&quot;The snapshot ‘tank/test@5? references only the new modified file ‘b’ and the newly created file ‘c’, and consumes no disk space.&quot;</description>
		<content:encoded><![CDATA[<p>Is &#8216;ZFS list&#8217; still reporting the same way in the current release (assuming you now add the required snapshots parameter)?  Maybe you ran into a minor stats display bug?  This (old) thread might be useful to you: <a href="http://www.mail-archive.com/zfs-discuss@opensolaris.org/msg05319.html" rel="nofollow">http://www.mail-archive.com/zfs-discuss@opensolaris.org/msg05319.html</a></p>
<p>Separately, I myself saw a minor bug in the wording of an error message when I was trying out ZFS on NexentaCore 1.0.  I ended up reporting it to the forums, OpenSolaris mailing list, and bugtracker.  In terms of triaging, little stats/text display bugs likely receive lower priority than most other bugs even though they may be low hanging fruit.  I also suspect these bugs are more noticeable to people new to and just learning ZFS rather than ZFS veterans who may not notice little inconsistencies like these.  I&#8217;ve run across a parallel situation in some university textbooks where there are internal errors and inconsistencies in example problems that make the problem hard to use or even useless as a learning example.  These issues usually don&#8217;t confound experienced people but they sure can throw a curveball for anyone new.</p>
<p>By the way, in your article above, you probably want to re-word this sentence which is self contradictory:<br />
&#8220;The snapshot ‘tank/test@3? references only the new modified file ‘b’ and consumes no disk space. Oh, but it seems it does consume some disk space! Why?&#8221;</p>
<p>This sentence appears to have a typo where 5 should be replaced with 4:<br />
&#8220;The snapshot ‘tank/test@5? references only the new modified file ‘b’ and the newly created file ‘c’, and consumes no disk space.&#8221;</p>
]]></content:encoded>
	</item>
</channel>
</rss>

