<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	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/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>Jump Networks Ltd</title>
	<atom:link href="http://www.jump.net.uk/feed" rel="self" type="application/rss+xml" />
	<link>http://www.jump.net.uk</link>
	<description>Bandwidth and Server Colocation in Telehouse North, London Docklands, UK</description>
	<lastBuildDate>Thu, 21 Mar 2013 01:43:56 +0000</lastBuildDate>
	<language>en</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.2</generator>
		<item>
		<title>Building energy efficient servers</title>
		<link>http://www.jump.net.uk/blog-energy-efficient-servers</link>
		<comments>http://www.jump.net.uk/blog-energy-efficient-servers#comments</comments>
		<pubDate>Fri, 02 Mar 2012 01:31:06 +0000</pubDate>
		<dc:creator>jamesr</dc:creator>
				<category><![CDATA[Uncategorized]]></category>

		<guid isPermaLink="false">http://www.jump.net.uk/?p=208</guid>
		<description><![CDATA[At Jump we really like energy efficient servers. We try to encourage our customers to choose power-friendly hardware, so much so that in 2008 we changed our pricing scheme to do away with a per U or per server charge, &#8230; <a href="http://www.jump.net.uk/blog-energy-efficient-servers">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>At Jump we really like energy efficient servers. We try to encourage our customers to choose power-friendly hardware, so much so that in 2008 we changed our pricing scheme to do away with a per U or per server charge, and charged purely based upon power usage (in VA, metered at idle).</p>
<p>Always on the lookout for well featured power friendly hardware, my current favourite picks are offerings from Supermicro &#8211; one can mix and match components so that hot swap drive bays are available even on very small systems, and the IPKVMS (Remote over IP Keyboard Video Mouse and Storage) feature only costs around an extra £15, compared to several hundred pounds from some vendors, and Intel NICs save frustrations with operating systems which don&#8217;t include binary firmware blobs in their default installers.</p>
<p>With criteria of IPKVMS, sandy bridge, hot swap drive bays, and 80plus gold or better PSU, the following options are all interesting.</p>
<p>Jump isn&#8217;t generally in the business of hardware building or sales, although we occasionally do do this for customers, so I&#8217;ve chosen to link to and give prices from http://www.lambda-tek.com , who we have no affiliation with, and the prices are just a snapshot as of 2012-03-02, but they offer mostly the best prices I&#8217;ve seen, and are able and happy to supply Supermicro kit without opening a trade account with them. Where they&#8217;re more than a few pounds more expensive than other suppliers for a given component I&#8217;ve mentioned it.</p>
<p>A build will consist of a chassis, motherboard, CPU, heatsink, RAM, and HDDs. I&#8217;ve not considered HDDs here, and I&#8217;ve only considered the X9SCL+-F motherboard, which features IPKVMS and two traditional Intel GigE ports (so no need for particularly new kernels to support the NICs). I&#8217;ve only considered the smallest and largest reasonable RAM options, it is possible to use 32GB RAM, but the price is astronomical right now.</p>
<p>Motherboard: £139.44 <a href="http://www.lambda-tek.com/~cs/B740738">X9SCL+-F</a><br />
Heatsink: £20.20 <a href="http://www.lambda-tek.com/~cs/B452212">SNK-P0046P</a> (£6 cheaper elsewhere)</p>
<p>CPU, one of: (Cores / Threads / Frequency / Max Turbo / Range)<br />
£24.85 <a href="http://www.lambda-tek.com/~cs/2351073">G460</a> (1/2/1.8/na Celeron)<br />
£31.14 <a href="http://www.lambda-tek.com/~cs/2288313">G530T</a> (2/2/2.0/na Celeron)<br />
£48.55 <a href="http://www.lambda-tek.com/~cs/2201622">G620T</a> (2/2/2.2/na Pentium)<br />
£46.10 <a href="http://www.lambda-tek.com/~cs/2288312">G630T</a> (2/2/2.3/na Pentium)<br />
£84.90 <a href="http://www.lambda-tek.com/~cs/2115527">i3-2100T</a> (2/4/2.5/na Core i3)<br />
£84.22 <a href="http://www.lambda-tek.com/~cs/2288311">i3-2120T</a> (2/4/2.6/na Core i3)<br />
£127.53 <a href="http://www.lambda-tek.com/~cs/2122400">E3-1220L</a> (2/4/2.2/3.4 Xeon E3)<br />
£189.03 <a href="http://www.lambda-tek.com/~cs/2115622">E3-1260L</a> (4/8/2.4/3.3 Xeon E3)</p>
<p>RAM, options include: (takes up to 4x unbuffered DIMMs)<br />
£8.81   <a href="http://www.lambda-tek.com/~cs/2107130">CT12872BA1339</a> (1x 1GB DIMM)<br />
£42.92  <a href="http://www.lambda-tek.com/~cs/B598984">CT2KIT51272BA1339</a> (2x 4GB DIMMs)</p>
<p>Chassis, one of: (* denotes Supermicro uses one of these chassis in Superserver with X9SCL motherboard / hot-swap drive bays / chassis depth / PSU)<br />
£110.02 <a href="http://www.lambda-tek.com/~cs/B765919">CSE-510T-203B</a> 2x 2.5&#8243; / 287mm / 80plus gold<br />
£196.95 <a href="http://www.lambda-tek.com/~cs/B545146">CSE-111LT-330CB</a> * 4x 2.5&#8243; / 558mm / 80plus gold (£30 cheaper elsewhere)<br />
£201.01 <a href="http://www.lambda-tek.com/~cs/B671118">CSE-113MTQ-330CB</a> 8x 2.5&#8243; / 508mm / 80plus gold<br />
£378.80 <a href="http://www.lambda-tek.com/~cs/B740519">CSE-113TQ-R500CB</a> 8x 2.5&#8243; / 597mm / 80plus platinum redundant<br />
£179.11 <a href="http://www.lambda-tek.com/~cs/B531365">CSE-813MTQ-350CB</a> * 4x 3.5&#8243; / 503mm / 80plus gold<br />
£343.55 <a href="http://www.lambda-tek.com/~cs/B671188">CSE-813MTQ-R400CB</a> * 4x 3.5&#8243; / 503mm / 80plus standard redundant</p>
<p>An entry level 510T-203B chassis, motherboard, heatsink, G460 CPU, 1GB RAM, comes to £303.32, just add disks.<br />
Higher end, 813MTQ-350CB, motherboard, heatsink, E3-1260L CPU, 16GB RAM, comes to £613.62, this, plus a pair of enterprise HDDs, is what we&#8217;re using for our <a href="http://www.jump.net.uk/dedicated">dedicated server</a> hardware &#8211; and meters at 39.6VA, 176mA@231V, 31.6W, PF0.78 (with no disks, booted into Ubuntu 10.10 server).</p>
<p>Edited 2012-03-06:<br />
Remove G440 CPU link (no support for speedstep so higher idle power)<br />
<a href="http://ark.intel.com/compare/63913,53415,53481,53484,53423,53427,53401,52275">comparison of above CPUs</a><br />
Added max turbo boost frequencies to CPU table</p>
<p>&nbsp;</p>
]]></content:encoded>
			<wfw:commentRss>http://www.jump.net.uk/blog-energy-efficient-servers/feed</wfw:commentRss>
		<slash:comments>1</slash:comments>
		</item>
		<item>
		<title>Configuring ZFS auto snapshot destroy thresholds</title>
		<link>http://www.jump.net.uk/blog-configuring-zfs-snapshot-destroy-thresholds</link>
		<comments>http://www.jump.net.uk/blog-configuring-zfs-snapshot-destroy-thresholds#comments</comments>
		<pubDate>Thu, 22 Sep 2011 09:50:38 +0000</pubDate>
		<dc:creator>jamesr</dc:creator>
				<category><![CDATA[Uncategorized]]></category>

		<guid isPermaLink="false">http://www.jump.net.uk/?p=181</guid>
		<description><![CDATA[A feature of ZFS automatic snapshots is automatic cleanup of snapshots when disk space is &#8216;low&#8217;. 95% full is considered an emergency, and all snapshots are purged, which probably isn&#8217;t desired with several TB free! The following will appear in &#8230; <a href="http://www.jump.net.uk/blog-configuring-zfs-snapshot-destroy-thresholds">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>A feature of ZFS automatic snapshots is automatic cleanup of snapshots when disk space is &#8216;low&#8217;. 95% full is considered an emergency, and all snapshots are purged, which probably isn&#8217;t desired with several TB free!</p>
<p>The following will appear in /var/adm/messages</p>
<pre>Sep 21 05:24:49 host time-sliderd: [ID 702911 daemon.crit] tank exceeded 90% capacity. Weekly, hourly and daily automatic snapshots were destroyed
Sep 21 05:24:49 host time-sliderd: [ID 702911 daemon.notice] 43 automatic snapshots were destroyed</pre>
<p>and later</p>
<pre>Sep 22 06:18:40 host time-sliderd: [ID 702911 daemon.emerg] tank is over 95% capacity. All automatic snapshots were destroyed
Sep 22 06:18:40 host time-sliderd: [ID 702911 daemon.notice] 758 automatic snapshots were destroyed</pre>
<p>Warning level exceeded: Destroy hourly and daily snapshots, oldest first, until threshold no longer exceeded.<br />
Critical level exceeded: Also destroy weekly snapshots, oldest first, until threshold no longer exceeded.<br />
Emergency level exceeded: Also destroy monthly snapshots, then frequent (15min) snapshots, until threshold no longer exceeded.</p>
<p>We can view the current thresholds as follows</p>
<pre>root@host:~# svcprop time-slider | grep level
zpool/critical-level integer 90
zpool/emergency-level integer 95
zpool/warning-level integer 80</pre>
<p>And change the thresholds as follows:</p>
<pre>svccfg -s time-slider setprop zpool/emergency-level=99
svccfg -s time-slider setprop zpool/critical-level=98
svccfg -s time-slider setprop zpool/warning-level=97
svcadm refresh time-slider</pre>
<p>References:<br />
<a href="http://arc.opensolaris.org/caselog/LSARC/2008/571/proposal.txt http://www.stormsail.com/2009/07/06/opensolaris-and-time-slider/">http://arc.opensolaris.org/caselog/LSARC/2008/571/proposal.txt</a><br />
<a href="http://www.stormsail.com/2009/07/06/opensolaris-and-time-slider/"> http://www.stormsail.com/2009/07/06/opensolaris-and-time-slider/</a></p>
]]></content:encoded>
			<wfw:commentRss>http://www.jump.net.uk/blog-configuring-zfs-snapshot-destroy-thresholds/feed</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>Configuring Cisco Easy VPN Server and Client on ASA 8.4 with Network Extension Mode and Split Tunneling</title>
		<link>http://www.jump.net.uk/blog-cisco-easy-vpn-on-asa</link>
		<comments>http://www.jump.net.uk/blog-cisco-easy-vpn-on-asa#comments</comments>
		<pubDate>Sun, 18 Sep 2011 23:36:48 +0000</pubDate>
		<dc:creator>jamesr</dc:creator>
				<category><![CDATA[Uncategorized]]></category>

		<guid isPermaLink="false">http://www.jump.net.uk/?p=158</guid>
		<description><![CDATA[This is an example of a clean Easy VPN (EzVPN) Server configuration with Network Extension Mode (NEM) and Split Tunneling, for Cisco ASA software version 8.4. The Cisco website has some more sample configurations, but they tend to be entire &#8230; <a href="http://www.jump.net.uk/blog-cisco-easy-vpn-on-asa">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>This is an example of a clean Easy VPN (EzVPN) Server configuration with Network Extension Mode (NEM) and Split Tunneling, for Cisco ASA software version 8.4. The Cisco website has some more sample configurations, but they tend to be entire device configurations, rather than just the changes that need applying from a clean or existing device configuration, and they haven&#8217;t been updated since the NAT configuration changed in ASA 8.3, which makes no nat configurations incompatible with earlier versions. </p>
<p>This has been tested on a pair of Cisco ASA devices running software 8.4(2), on the ASA5505 with base licence (part number ASA5505-BUN-K9, available for about £185+VAT new from reputable places).</p>
<p>Sections in <strong>bold</strong> are ones you will almost certainly need / want to change for your specific configuration.<br />
Sections in <em>italics</em> show that the name has been chosen by myself, and could be changed in your configuration, as long as you&#8217;re consistent and change every occurrence of that name.</p>
<pre>EASY VPN SERVER SIDE
====================

!=== A clean start ===
configure factory-default <strong>192.168.80.1 255.255.255.0</strong>

!=== Generic Basic config ===
hostname <strong>easyserver</strong>
interface Vlan2
 ip address <strong>192.0.2.2 255.255.255.0</strong>
route outside 0.0.0.0 0.0.0.0 <strong>192.0.2.1</strong> 1
!-- Optional - allow pings outbound
policy-map global_policy
  class inspection_default
   inspect icmp

!=== IPsec configuration ===
crypto ipsec ikev1 transform-set <em>MapOutsideDynamicXform</em> esp-aes-256 esp-sha-hmac
crypto dynamic-map <em>MapOutsideDynamic</em> 5 set ikev1 transform-set <em>MapOutsideDynamicXform</em>
crypto map <em>MapOutside</em> 60 ipsec-isakmp dynamic <em>MapOutsideDynamic</em>
crypto map <em>MapOutside</em> interface outside
crypto ikev1 enable outside

!=== Group / tunnel policy and logins ===
group-policy <em>easyvpnclientpolicy</em> internal
group-policy <em>easyvpnclientpolicy</em> attributes
 nem enable
username <strong>eznemuser1</strong> password <strong>eznemuser1pass</strong>
tunnel-group <strong>eznemgroup</strong> type remote-access
tunnel-group <strong>eznemgroup</strong> general-attributes
 default-group-policy <em>easyvpnclientpolicy</em>
tunnel-group <strong>eznemgroup</strong> ipsec-attributes
 ikev1 pre-shared-key <strong>eznemgrouppass</strong>

!=== No NAT within our own extended network ===
!-- This is necessary otherwise the default PAT rule will cause
!-- RPF failures and easyclient network won't be able to access
!-- the easyserver network, and easyserver network access to
!-- easyclient will be subject to PAT.
object network <em>clientnet</em>
 subnet <strong>10.67.66.0 255.255.255.0</strong>
object network <em>servernet</em>
 subnet <strong>192.168.80.0 255.255.255.0</strong>
nat (inside,outside) source static <em>servernet servernet</em> destination static <em>clientnet clientnet</em> no-proxy-arp route-lookup

!=== For tunnel all clients to obtain Internet access ===
!-- If you want to tunnel everything, use this section, not the
!-- one below. This might be useful if the easyclient is on
!-- an ISP which tampers with traffic to the internet, but the
!-- easyserver is on more desirable connectivity.
object network <em>clientnet</em>
 nat (outside,outside) dynamic interface
same-security-traffic permit intra-interface

!=== For split tunneling ===
!-- You don't need the section above if you use this one. It
!-- causes only traffic to the networks listed below to go
!-- over the VPN, other traffic to the Internet uses the local
!-- ISP of the easyclient.
group-policy <em>easyvpnclientpolicy</em> attributes
 split-tunnel-policy tunnelspecified
 split-tunnel-network-list value <em>SplitTunnelNetworks</em>
access-list <em>SplitTunnelNetworks</em> standard permit <strong>10.0.0.0 255.0.0.0</strong>
access-list <em>SplitTunnelNetworks</em> standard permit <strong>172.16.0.0 255.240.0.0</strong>
access-list <em>SplitTunnelNetworks</em> standard permit <strong>192.168.0.0 255.255.0.0</strong></pre>
<p>In contrast, the client side is much easier</p>
<pre>EASY VPN CLIENT SIDE
====================

!=== A clean start ===
configure factory-default <strong>10.67.66.1 255.255.255.0</strong>

!=== Generic Basic config ===
hostname <strong>easyclient</strong>
interface Vlan2
 ip address <strong>198.51.100.2 255.255.255.0</strong>
route outside 0.0.0.0 0.0.0.0 <strong>198.51.100.1</strong> 1

!=== Easy VPN Client ===
vpnclient server <strong>192.0.2.2</strong>
vpnclient mode network-extension-mode
vpnclient nem-st-autoconnect
vpnclient vpngroup <strong>eznemgroup</strong> password <strong>eznemgrouppass</strong>
vpnclient username <strong>eznemuser1</strong> password <strong>eznemuser1pass</strong>
vpnclient enable</pre>
]]></content:encoded>
			<wfw:commentRss>http://www.jump.net.uk/blog-cisco-easy-vpn-on-asa/feed</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>VPS disk space price reduction</title>
		<link>http://www.jump.net.uk/news-vps-disk-space-price-reduction</link>
		<comments>http://www.jump.net.uk/news-vps-disk-space-price-reduction#comments</comments>
		<pubDate>Sun, 05 Sep 2010 16:34:06 +0000</pubDate>
		<dc:creator>admin</dc:creator>
				<category><![CDATA[Uncategorized]]></category>

		<guid isPermaLink="false">http://www.jump.net.uk/?p=129</guid>
		<description><![CDATA[We&#8217;ve reduced the cost of additional RAID1 iSCSI backed storage space for virtual private servers from £30/year per 20GB, to £10/year per 10GB.]]></description>
			<content:encoded><![CDATA[<p>We&#8217;ve reduced the cost of additional RAID1 iSCSI backed storage space for virtual private servers from £30/year per 20GB, to £10/year per 10GB.</p>
]]></content:encoded>
			<wfw:commentRss>http://www.jump.net.uk/news-vps-disk-space-price-reduction/feed</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
		<item>
		<title>Now peering at LINX</title>
		<link>http://www.jump.net.uk/news-now-peering-at-linx</link>
		<comments>http://www.jump.net.uk/news-now-peering-at-linx#comments</comments>
		<pubDate>Fri, 25 Jun 2010 17:46:18 +0000</pubDate>
		<dc:creator>admin</dc:creator>
				<category><![CDATA[Uncategorized]]></category>

		<guid isPermaLink="false">http://www.jump.net.uk/?p=92</guid>
		<description><![CDATA[Jump Networks is delighted to be peering at LINX, with a GE connection to the LINX Brocade LAN.]]></description>
			<content:encoded><![CDATA[<p>Jump Networks is delighted to be peering at LINX, with a GE connection  to the LINX Brocade LAN.</p>
]]></content:encoded>
			<wfw:commentRss>http://www.jump.net.uk/news-now-peering-at-linx/feed</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
	</channel>
</rss>
