<?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: VMware ESX, EMC CLARiiON Arrays, and Multiple Protocols</title>
	<atom:link href="http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/</link>
	<description>The weblog of an IT pro specializing in virtualization, storage, and servers</description>
	<lastBuildDate>Wed, 08 Feb 2012 17:13:47 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=</generator>
	<item>
		<title>By: slowe</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-48583</link>
		<dc:creator>slowe</dc:creator>
		<pubDate>Thu, 10 Jun 2010 08:22:21 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-48583</guid>
		<description>Mark,

My question is why do you want to have this setup? What is driving you to say that one LUN should be visible on one HBA and a different LUN visible on a different HBA? Depending on why you want to have this sort of design, there might be a better way of approaching the problem.</description>
		<content:encoded><![CDATA[<p>Mark,</p>
<p>My question is why do you want to have this setup? What is driving you to say that one LUN should be visible on one HBA and a different LUN visible on a different HBA? Depending on why you want to have this sort of design, there might be a better way of approaching the problem.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mark</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-48580</link>
		<dc:creator>Mark</dc:creator>
		<pubDate>Thu, 10 Jun 2010 04:54:16 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-48580</guid>
		<description>Hi,
I am having the same problem. I have 1 host with 2 x HBA ports. I want to access Lun0 down path 1 and Lun1 down path 2. Now when my host registers with the clariion it adds both pwwns to the same host name as it recognizes the same fqdn and IP address. So now I have this Host1 &gt;pwwn1 and pwwn2.
Now I cannot separate access down each path to the luns as I only have 1 host with 2 x pwwns under it. I want 2 hosts with 1 pwwn under each so I can create two storage groups with one pwwn under each.
How can this be done? by manual registration and fake IP addresses only?</description>
		<content:encoded><![CDATA[<p>Hi,<br />
I am having the same problem. I have 1 host with 2 x HBA ports. I want to access Lun0 down path 1 and Lun1 down path 2. Now when my host registers with the clariion it adds both pwwns to the same host name as it recognizes the same fqdn and IP address. So now I have this Host1 &gt;pwwn1 and pwwn2.<br />
Now I cannot separate access down each path to the luns as I only have 1 host with 2 x pwwns under it. I want 2 hosts with 1 pwwn under each so I can create two storage groups with one pwwn under each.<br />
How can this be done? by manual registration and fake IP addresses only?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: slowe</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47782</link>
		<dc:creator>slowe</dc:creator>
		<pubDate>Thu, 01 Apr 2010 13:21:09 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47782</guid>
		<description>InsaneGeek,

Small clarification: it&#039;s not that FC and iSCSI on the same host isn&#039;t supported; you could do FC to one array and iSCSI to a separate array with no problems. It&#039;s FC and iSCSI on the same host to the same array that isn&#039;t supported by EMC, and FC and iSCSI to the same LUN simultaneously that isn&#039;t supported by VMware.

As for a cluster, FC on some hosts and iSCSI on some hosts to the same array should be fine. Not necessarily a recommended configuration, but it should work.</description>
		<content:encoded><![CDATA[<p>InsaneGeek,</p>
<p>Small clarification: it&#8217;s not that FC and iSCSI on the same host isn&#8217;t supported; you could do FC to one array and iSCSI to a separate array with no problems. It&#8217;s FC and iSCSI on the same host to the same array that isn&#8217;t supported by EMC, and FC and iSCSI to the same LUN simultaneously that isn&#8217;t supported by VMware.</p>
<p>As for a cluster, FC on some hosts and iSCSI on some hosts to the same array should be fine. Not necessarily a recommended configuration, but it should work.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: InsaneGeek</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47780</link>
		<dc:creator>InsaneGeek</dc:creator>
		<pubDate>Thu, 01 Apr 2010 05:05:05 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47780</guid>
		<description>So, FC &amp; iscsi the same host is unsupported... any idea if having some hosts using FC &amp; some using iscsi to the same lun is supported in a HA cluster (a host is either all FC or iscsi not both)?</description>
		<content:encoded><![CDATA[<p>So, FC &amp; iscsi the same host is unsupported&#8230; any idea if having some hosts using FC &amp; some using iscsi to the same lun is supported in a HA cluster (a host is either all FC or iscsi not both)?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Hesse</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47757</link>
		<dc:creator>David Hesse</dc:creator>
		<pubDate>Fri, 26 Mar 2010 10:30:10 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47757</guid>
		<description>very good article. Recently I had a discussion with one of my colleagues about that exact topic and we had different opionions on what is possible and what not. Your article has tackled the mystery. Thank you :-)</description>
		<content:encoded><![CDATA[<p>very good article. Recently I had a discussion with one of my colleagues about that exact topic and we had different opionions on what is possible and what not. Your article has tackled the mystery. Thank you <img src='http://blog.scottlowe.org/wp-includes/images/smilies/icon_smile.gif' alt=':-)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: slowe</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47721</link>
		<dc:creator>slowe</dc:creator>
		<pubDate>Sat, 20 Mar 2010 04:28:23 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47721</guid>
		<description>John Philips,

It&#039;s absolutely possible, as you&#039;ve so clearly discovered. It&#039;s just not SUPPORTED. VMware does not support accessing the same LUN over both FC and iSCSI.</description>
		<content:encoded><![CDATA[<p>John Philips,</p>
<p>It&#8217;s absolutely possible, as you&#8217;ve so clearly discovered. It&#8217;s just not SUPPORTED. VMware does not support accessing the same LUN over both FC and iSCSI.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: John Philips</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47719</link>
		<dc:creator>John Philips</dc:creator>
		<pubDate>Fri, 19 Mar 2010 16:56:14 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47719</guid>
		<description>I have several ESXi 4.0 hosts connected to a CX4 SAN.  Some hosts connect with iSCSI, some with FC, some with both.  For the hosts with both, the &quot;Manage Paths...&quot; portion of the vCenter GUI shows that the same datastores are accessible via iSCSI and FC paths.  The SAN is still running v28 software.

Maybe it&#039;s not possible to use multipath/load balancing over iSCSI &amp; FC at the same time, but at least for failover purposes it looks do-able out-of-the-box with no special configuration.</description>
		<content:encoded><![CDATA[<p>I have several ESXi 4.0 hosts connected to a CX4 SAN.  Some hosts connect with iSCSI, some with FC, some with both.  For the hosts with both, the &#8220;Manage Paths&#8230;&#8221; portion of the vCenter GUI shows that the same datastores are accessible via iSCSI and FC paths.  The SAN is still running v28 software.</p>
<p>Maybe it&#8217;s not possible to use multipath/load balancing over iSCSI &amp; FC at the same time, but at least for failover purposes it looks do-able out-of-the-box with no special configuration.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Vinod Pant</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47684</link>
		<dc:creator>Vinod Pant</dc:creator>
		<pubDate>Thu, 11 Mar 2010 11:51:45 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47684</guid>
		<description>I am trying to create multipathing between Clariion CX3-10 and ESX4, but have not got the success so far. It detects the path 5 which is correct but shows only one path under &quot;managed path&quot;. What I did so far is, create multiple vmkernel port groups, assigned each port group a dedicated NIC and then bind these vmk port groups on the console.
I am using software iSCSI intiator at ESX4 and iSCSI connection from CX3-10.

Is there any doc available in the internet that could explained the process? I have not found any.
Thanks,
V</description>
		<content:encoded><![CDATA[<p>I am trying to create multipathing between Clariion CX3-10 and ESX4, but have not got the success so far. It detects the path 5 which is correct but shows only one path under &#8220;managed path&#8221;. What I did so far is, create multiple vmkernel port groups, assigned each port group a dedicated NIC and then bind these vmk port groups on the console.<br />
I am using software iSCSI intiator at ESX4 and iSCSI connection from CX3-10.</p>
<p>Is there any doc available in the internet that could explained the process? I have not found any.<br />
Thanks,<br />
V</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Evaldas</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47682</link>
		<dc:creator>Evaldas</dc:creator>
		<pubDate>Wed, 10 Mar 2010 16:20:08 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47682</guid>
		<description>Scott, we actually use the same LUN via multiprotocol-multipath to ESX from OpenSolaris COMSTAR targets, FC and iSCSI (1G/10G), basically for failover, I think you could call EMC&#039;s host registration as &quot;views&quot; in COMSTAR, there you can take initiator identifiers wwn.., iqn.. and put them in a group and then create a view for that group on a particual LUN (Mike LaSpina has a described it in very nice details here: http://blog.laspina.ca/ubiquitous/multi_protocol_storage_provisioning_with ). Here is &quot;Manage paths&quot; screenshot for one of our LUNs - http://zaibasweb.nerim.net/esx-multiprotocol-multipath.png. The only annoyance with this setup is that VC &quot;Storage Views&quot; tab is not working. Do you have an idea if VMware has any plans on supporting multiproto for same LUN, working in that direction ?</description>
		<content:encoded><![CDATA[<p>Scott, we actually use the same LUN via multiprotocol-multipath to ESX from OpenSolaris COMSTAR targets, FC and iSCSI (1G/10G), basically for failover, I think you could call EMC&#8217;s host registration as &#8220;views&#8221; in COMSTAR, there you can take initiator identifiers wwn.., iqn.. and put them in a group and then create a view for that group on a particual LUN (Mike LaSpina has a described it in very nice details here: <a href="http://blog.laspina.ca/ubiquitous/multi_protocol_storage_provisioning_with" rel="nofollow">http://blog.laspina.ca/ubiquitous/multi_protocol_storage_provisioning_with</a> ). Here is &#8220;Manage paths&#8221; screenshot for one of our LUNs &#8211; <a href="http://zaibasweb.nerim.net/esx-multiprotocol-multipath.png" rel="nofollow">http://zaibasweb.nerim.net/esx-multiprotocol-multipath.png</a>. The only annoyance with this setup is that VC &#8220;Storage Views&#8221; tab is not working. Do you have an idea if VMware has any plans on supporting multiproto for same LUN, working in that direction ?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Vijay Swami</title>
		<link>http://blog.scottlowe.org/2010/03/05/vmware-esx-emc-clariion-arrays-and-multiple-protocols/comment-page-1/#comment-47660</link>
		<dc:creator>Vijay Swami</dc:creator>
		<pubDate>Tue, 09 Mar 2010 04:32:02 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/?p=1856#comment-47660</guid>
		<description>Scott, what *specific* use cases were you think where this would be useful? I&#039;m trying to think of any where it would be handy to serve iSCSI and FC from the a single ESX host but can&#039;t think of any, nor can recall any with current/previous customer&#039;s where it would have came in handy. Can you elaborate on some specific scenarios you had in mind? Typically customer&#039;s like to segment off test/dev environments into their own HA/DRS clusters so they are completely fenced from production. Even with the advent of resource pools, I see advantages to a separate dev/test cluster in many customer environments. So my point is, while I think it&#039;s a nice feature, I just can&#039;t seem to come up with any real world applications. 

Thoughts?</description>
		<content:encoded><![CDATA[<p>Scott, what *specific* use cases were you think where this would be useful? I&#8217;m trying to think of any where it would be handy to serve iSCSI and FC from the a single ESX host but can&#8217;t think of any, nor can recall any with current/previous customer&#8217;s where it would have came in handy. Can you elaborate on some specific scenarios you had in mind? Typically customer&#8217;s like to segment off test/dev environments into their own HA/DRS clusters so they are completely fenced from production. Even with the advent of resource pools, I see advantages to a separate dev/test cluster in many customer environments. So my point is, while I think it&#8217;s a nice feature, I just can&#8217;t seem to come up with any real world applications. </p>
<p>Thoughts?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

