First of all we need to get the details of our current swap setup:

bash-3.00# swap -l
swapfile             dev  swaplo blocks   free
/dev/zvol/dsk/rpool/swap 256,2      16 4194288 4194288

New step is to increase the size of the ZFS ‘filesystem’ under the root pool (here called the default, rpool).

bash-3.00# zfs set volsize=4G rpool/swap

Once the filesystem size has been increased, we need to actually add it as swap. The normal swap command will do this - we just need to make sure we’re pointing it at the correct ZFS device:

bash-3.00# env NOINUSE_CHECK=1 swap -a /dev/zvol/dsk/rpool/swap $((8+4194288))

Let’s just check the status via ZFS:

bash-3.00# zfs list rpool/swap
NAME         USED  AVAIL  REFER  MOUNTPOINT
rpool/swap     4G  3.16G  2.03G  -

And finally we can see the new swap space we’ve just added:

bash-3.00# swap -l
swapfile             dev  swaplo blocks   free
/dev/zvol/dsk/rpool/swap 256,2      16 4194288 4194288
/dev/zvol/dsk/rpool/swap 256,2  4194304 4194304 4194304

A simple handful of commands, and no downtime - adding extra swap space using ZFS on Solaris is pretty painless. In another post I’ll explore how to grow ZFS filesystems like /var.

Browse Related Posts

• Sun and Oracle - aftermath of the big event
• Playing with Solaris processor sets
• Easy Solaris log file management with logadm
• Solaris basics: checking which package a file belongs to
• Solaris basics: configuring the DHCP client

If those two names don’t have you instantly clicking links, shame on you. Steven Furber was one of the original team at Acorn who came up with the legendary BBC Micro, and who went on to design the Acorn RISC Machine processor.

David Brown worked with Andy Bechtolsheim to design the original Sun workstation, and was one of the ‘dream team’ who worked with Jim Clarke to form Silicon Graphics. These guys really are rockstars of the computing world.

The interview covers the early adventures of the BBC Micro, how that lead on to the development of the ARM CPU, and the design decisions needed to build power-efficient processors.

Energy efficient computing is becoming a really hot topic (if you’ll pardon the pun) and I think it would be excellent to see the legacy of Acorn live on if ARM processors start making inroads to the data centre. With OpenSolaris being ported to the latest ARM Cortex designs this could be happening sooner than we think.

Read the full article at http://queue.acm.org/detail.cfm?id=1716385

Similar Posts

• OpenDS 2.0 is now out
• Mad Larry seeks to rule all with Sun/Oracle
• New home for orphaned Sun projects
• Using wget to download files from Sun Download Centre
• Silly SAN tricks - finding the WWN of an HBA from Solaris

And that’s the problem - there are a whole raft of command line options, and if you’re just starting out you’re presented with a dizzying array of possible optimisations.

Theirry’s post discusses some of the obvious ‘quick win’s you can make, as well as covering their potential downsides. It covers such gems as -fast:

If you are in a rush, you can use the -fast option. What it really does is triggering a set of other options for maximum runtime performance.

Head on over to this page and have a read through the full post - it’s very handy and has some useful tips on building some really optimised code.

Browse Related Posts

• New home for orphaned Sun projects
• Sun and Oracle - aftermath of the big event
• SuCLE lives: Oracle/Sun merger approved by the EU
• Optimising performance for parallel processing
• Sun Grid Engine for Dummies

Izumo Shinohara, a recently ex Sun employee, has setup a new site - Red Giant Phase - which aims to provide a new home for these orphaned projects.

Currently listed are projects like Wonderland, Dark Star, and Project DReaM, amongst others.

It’ll be interesting to see what happens to these open sourced projects now that they’re outside Sun - I wish them all the best.

Browse Related Posts

• Sun Studio compiler options - a beginners guide
• Sun and Oracle - aftermath of the big event
• SuCLE lives: Oracle/Sun merger approved by the EU
• Optimising performance for parallel processing
• Sun Grid Engine for Dummies

AI replaces Jumpstart for network installs. If you’re deploying OpenSolaris from scratch, this is fine - build one machine manually, set it up as an AI server, and roll out the rest. If you’ve got an existing Solaris infrastructure, however, this becomes a pain of a pain. An additional issue to take into account is that AI can only deploy OpenSolaris to SPARC machines which support WANBoot - so you’ll need to check your OBP versions.

This is the current matrix of what can be installed and how:

I’m sure lots of people are in no hurry to migrate their Jumpstart server to OpenSolaris - especially as it’s probably a SPARC box.

The big advantage of AI is that it’s very easy to get going. I’m going to work through deploying AI to demonstrate how simple it is. In this environment, the OpenSolaris machine that will be used as an AI server is also my workstation. I’ve got a Netra T1 running a complex Jumpstart + JET setup with lots of customisations - I don’t want to replace that, but I do want to use it’s DHCP server.

AI is managed via the installadm tool. It’s probably not installed on your OpenSolaris machine by default, so you’ll need to add it:

pfexec pkg install SUNWinstalladm-tools

Once you’ve got the tools in place, you need to setup the install server. Download an OpenSolaris image - but be careful! The Live CD ISO cannot be used to setup an AI server, you have to download the AI version of the ISO.

Head on over to http://hub.opensolaris.org/bin/view/Main/downloads to grab the relevant ISO.

Once you’ve got the ISO, you can setup the install server. It works based on services - each OS release for each platform is treated as a different service. The you add clients, and tell the client which service it will use to boot from.

I’m going to be sticking AI under /export - the traditional place in Solaris for shared filesystems. I just want to create one install service for OpenSolaris 06/09 x86, which I’ll call 0609×86.

The full command line is:

installadm create-service -n (service_name) \
	-s (source_AI_ISO) (AI_service_data_directory)

Here’s the full command line along with the output:

root@grond:/export# /usr/sbin/installadm create-service -n 0609x86 \
	-s /export/torrents/osol-0906-ai-x86.iso \
	/export/aiserver/osol-0906-ai-x86
Setting up the target image at /export/aiserver/osol-0906-ai-x86 ...
Registering the service 0609x86._OSInstall._tcp.local

Detected that DHCP is not set up on this server.
If not already configured, please create a DHCP macro
named dhcp_macro_0609x86 with:
   Boot server IP (BootSrvA) : 192.168.13.100
   Boot file      (BootFile) : 0609x86
   GRUB Menu      (GrubMenu) : menu.lst.0609x86
If you are running Sun's DHCP server, use the following
command to add the DHCP macro, dhcp_macro_0609x86:
   /usr/sbin/dhtadm -g -A -m dhcp_macro_0609x86 -d :BootSrvA=192.168.13.100: \
	BootFile=0609x86:GrubMenu=menu.lst.0609x86:

Additionally, if the site specific symbol GrubMenu
is not present, please add it as follows:
   /usr/sbin/dhtadm -g -A -s GrubMenu -d Site,150,ASCII,1,0

Note: Be sure to assign client IP address(es) if needed
(e.g., if running Sun's DHCP server, run pntadm(1M)).
Service discovery fallback mechanism set up

Helpfully, installadm tells us what commands to run on our DHCP server. First we’ll need to add the GrubMenu symbol (it won’t exist by default) and then we can add in the DHCP macro for the service. Just copy and paste the two commands on your Jumpstart server.

With that out of the way, we can now setup a client. In this case, I have a Sun v20z with a MAC address of 00:09:3d:12:ff:80 on bge0.

We need to run installadm to create the client, giving it the MAC address and telling it which install service to use. The command line is:

installadm create-client -e (MAC_address) -n (AI_service_name_to_use) \
	-t (AI_service_data_directory)

Here’s the full command line with the output:

root@grond:/export# /usr/sbin/installadm create-client \
	-e 00:09:3d:12:ff:80 -n 0609x86 \
	-t /export/aiserver/osol-0906-ai-x86
Setting up X86 client...
Service discovery fallback mechanism set up

Detected that DHCP is not set up on this server.
If not already configured, please create a DHCP macro
named 0100093D12FF80 with:
   Boot server IP (BootSrvA) : 192.168.13.100
   Boot file      (BootFile) : 0100093D12FF80
If you are running Sun's DHCP server, use the following
command to add the DHCP macro, 0100093D12FF80:
   /usr/sbin/dhtadm -g -A -m 0100093D12FF80 -d :BootSrvA=192.168.13.100: \
	BootFile=0100093D12FF80:GrubMenu=menu.lst.0100093D12FF80:

Note: Be sure to assign client IP address(es) if needed
(e.g., if running Sun's DHCP server, run pntadm(1M)).

Once again installadm will helpfully tell us what commands we need to run on our DHCP server to add the macros for this client.

Over on the Jumpstart server, here’s the output of dhtadm showing us the configured macros on the Sun DHCP server (with some line breaks to make it a bit more readable):

bash-3.00# dhtadm -P
Name                    Type            Value
==================================================
dhcp_macro_0609x86      Macro           :BootSrvA=192.168.13.100: \
	BootFile=0609x86:GrubMenu=menu.lst.0609x86:
0100093D12FF80          Macro           :BootSrvA=192.168.13.100: \
	BootFile=0100093D12FF80:GrubMenu=menu.lst.0100093D12FF80:
v20z                    Macro           :BootFile=0100093D12FF80: \
	BootSrvA=192.168.13.101:
192.168.13.0
GrubMenu                Symbol          Site,150,ASCII,1,0

I’ve removed all of the other stuff that Jumpstart puts in there to clearly show the AI macros that have been added.

At this stage, we can just SSH into the V20z’s ILOM, power on the chassis, and go into the BIOS to change the boot order. PXEboot will then send out a DHCP request, and we’ll then see the OpenSolaris grub menu.

From that point onwards it’s a hands-off install. For more details on the entire process, have a read through the OpenSolaris Automated Installer Guide.

Having played around with AI for a bit now, I’m not that impressed to be honest. I can see that it could be easier for new users who’ve never touched Solaris before - as you can see, it doesn’t take much to setup an install server and configure clients.

However, there’s a big installed base of Solaris users out there, and they’ve all got Jumpstart. AI lacks the features, flexibility and power of Jumpstart - it’s not ready as a replacement just yet. So being forced to use it to be able to deploy OpenSolaris just means many existing Solaris shops won’t bother - integration with Jumpstart for OpenSolaris could well speed up it’s acceptance and adoption.

With so many Solaris users out there I think that OpenSolaris needs a lot of work to become a credible upgrade or migration path. Both AI and the new IPS packaging system show promise, but they’re a long way from being usable replacements to existing Solaris technologies.

Browse Related Posts

• Sun and Oracle - aftermath of the big event
• Configuring Sun’s Extras repository in OpenSolaris
• A quick and easy start to HPC development
• A review of the Sun Ultra 24 workstation
• Easy imaging of Solaris using flars - Solaris Flash Archives

The webcast held no surprises, really. As I mentioned previously about the Sun/Oracle merger and Larry’s talk on Oracle’s use of Sun technology, Oracle weren’t going to ditch Sun’s hardware line. The analysts were full of hysterics and gloom, but I’ve yet to meet an analyst who has the slightest clue of what’s going on. They’re paid to make noise and sell ‘research’, not to know what they’re talking about.

As predicted, there’s more investment in Sun’s hardware line, including lots more tasty new CMT processors, and a scaling up of the line to larger multi-socket machines. The high end gear will continue, as will the partnership with Fujitsu. SPARC continues to get a lot of investment and love, and will be a big focus going forwards. Amen to that.

Pretty much all of the software stack will stay and get integrated with Oracle’s offerings. I note with distaste that Oracle’s crappy Internet Directory remains the ‘enterprise’ offering for LDAP and identity management, with Sun’s LDAP products being pushed at smaller deployments. On the OS side, it’s a bit of Linux, and Solaris, Solaris, Solaris - Oracle recognise it’s the best commercial UNIX currently on the market, and that the feature set is unmatched.

Storage lives on, with Sun’s excellent Amber Road Storge 7000 Unified Storage boxes becoming ZFS appliances. Particularly exciting is their integration into OEM - imagine simple management of RMAN backups to ZFS appliances, giving low level snapshots and all sorts of goodness. I can see a lot of places going for that in a big way.

The big question for me was around OpenSolaris. No mention of it at all. It’s Open Source - that particular cat is out of the bag, and it’s not going away. So the question is what sort of effort will Oracle put behind it? Lots of new, OpenSolaris specific features - the new IPS packaging system and the Automated Installer - have potential, but aren’t up to scratch yet, and quite frankly don’t play well with existing Solaris infrastructure.

My bet is we’ll see less effort in re-inventing the wheel, and more focus on making OpenSolaris a more palatable Solaris 11. There’s a big Solaris installed base out there, and the focus on x86 and new features has so far meant that OpenSolaris isn’t really a credible upgrade path.

As I expected when I first heard the news, Oracle are going to be leveraging Sun’s technology and services and own and optimise the entire stack, from the silicon up to the application. This gives them a chance to really tune everything and to go head to head with IBM. May you live in interesting times, as they say.

Obviously there’s more, lots more. Oracle have handily posted up each section of the webcast so you can pick and choose which session you want to watch here. There are also a series of special short webcasts which focus on specific product areas - you can view them all here.

PS: as a side note, Thomas Kurian, who presented the Software Strategy webcast, managed to give one of the dullest presentations I’ve seen. Seriously, that was a really important session, but I almost nodded off a couple of times. Dire.

Browse Related Posts

• New SPARC solutions are on their way
• SuCLE lives: Oracle/Sun merger approved by the EU
• Mad Larry seeks to rule all with Sun/Oracle
• Larry Ellison on Cloud Computing
• Oracle have bought Sun

BIND logging is managed in /etc/named.conf and is split into two parts.

First of all, configure BIND to store PIDs, statistics files etc. in a dedicated directory. In this case, I’m using /var/named/log:

options {
        directory "/var/named/log";
        pid-file "/var/named/log/named.pid";
        dump-file "/var/named/log/named_dump.db";
        memstatistics-file "/var/named/log/named.memstats";
        statistics-file "/var/named/log/named.stats";
};

That’s all pretty straightforward. Armed with this we can write some scripts to plug BIND statistics into RRDTool or similar, and get some nice graphs of performance.

Next, we can use the logging directive to configure log files and the streams of events we want to log:

logging {
        channel default_log {
                file "/var/named/log/named.log" versions 3 size 10m;
                print-time yes;
                print-category yes;
                severity info;
                };
        channel security_log {
                file "/var/named/log/named.security.log" versions 3 size 5m;
                print-time yes;
                print-category yes;
                severity notice;
                };
        category default { default_log; };
        category security { security_log; };
        category lame-servers { null; };
};

You can see I’ve configured a default log of INFO events, which will by cycled when it reaches 10mb, with 3 copies being kept. I’m also logging security events with a priority of NOTICE and higher - 3 versions of this are kept, and the security log gets cycled when it reaches 5mb.

For all the log events I want to print the time stamp and the category - without these it can be difficult to work out what happened and when, which makes the logs pretty much worthless.

The syntax to configure logging in BIND is pretty straightforward and is very similar to how you’d configure your zones in named.conf.

If you want to be slack you can just cut and paste this into /etc/named.conf and restart BIND - instant logging. I run BIND under a non-root user, and the BIND logging directory is owned by that user and group and locked down - I’d suggest doing the same at a minimum.

Browse Related Posts

• A Rough Guide to LDAP

On the more interesting ones is by of the presentation by Yan Fisher, who is Benchmark Lead in Sun’s Technical Marketing Systems Group. His presentation is an update on benchmarking in HPC.

Head on over to Sun’s HPC Watercooler to watch it.

Browse Related Posts

• Optimising performance for parallel processing
• Sun HPC Consortium presentations and videos posted
• Interview with LANL researcher about using GPUs
• Sun’s research report on Hardware Transactional Memory
• New HPC for Dummies book announced

Oracle are holding a webcast next Wednesday 27th, where Mad Larry will be laying out his stalls and plotting the roadmap to world domination. You can sign up here - well worth a listen, if only because now Oracle are fully off the leash they’re free to really put the boot in to IBM and HP.

On a related note, I find Monty Widenius’ objections to the merger/takeover/sale bizarre. Sun paid the shareholders a cool $1 billion for MySQL AB - a ridiculous amount. They can do what they want with it. Surely if that caused you problems, you shouldn’t have sold it in the first place?

Selling something, then trying to force the new owners to let you have back control so you can build a competing commercial business off it - for free - is, quite frankly, greedy and deeply shady. And in the meantime, the delay has damaged Sun, their customers, partners - oh, and the career prospects of all those MySQL AB guys who now work for Sun. I’m sure Monty will be crying them all a river as he rolls around in the big stack of cash he got from Sun.

It’s a strange contrast to the behaviour of the Jboss folks, where after the sale the application server has been transformed via RedHat cash into a credible platform that’s met with some solid commercial success.

Browse Related Posts

• Sun and Oracle - aftermath of the big event
• Mad Larry seeks to rule all with Sun/Oracle
• New SPARC solutions are on their way
• Larry Ellison on Cloud Computing
• Oracle have bought Sun

Which makes it all the more important that I can work with IT managers and CTOs who don’t just ‘get’ technology, but are also accomplished and skilled managers. I’ve worked with a number of people who excel at those roles (to the immense benefit of their business).

One of the best is someone I worked with at a dot-com a decade ago. The planets have aligned and we’re working together again at the moment, on some complex IT issues that are just as interesting as they were years ago.

This time round, though, he’s blogging about IT management issues (and online gambling). Head on over to Transmissions From a Free Ranging Agent of Kaos for a read - it’s informative and useful stuff.

Similar Posts

• Solaris 9 can’t import it’s SVM metasets when booting
• NVidia: Mad Science
• Installing OpenSolaris with the Automated Installer
• Finding the WWN in Solaris followup - making it easier
• Solaris basics: configuring the DHCP client