Ram's Live Journal

Compiling Ruby 1.9.3-p0 on Belenix

dynamicproxy — Tue, 01 Nov 2011 18:50:39 GMT

After a few exciting weeks at work, I picked up Belenix work again today. I decided to give Ruby 1.9.3-p0 a try, since this is the latest stable.

I ran into a libelf issue, where my the default libelf that came with illumos didn't support large file sizes. People have faced this before already,

http://www.ruby-forum.com/topic/2689222
http://redmine.ruby-lang.org/issues/5384

I was able to address this by installing a different libelf from here http://www.mr511.de/software/ with a prefix of /usr/local

CFLAGS="-I/usr/local/include/libelf -L/usr/local/include" LDFLAGS="-L/usr/local/lib -R/usr/local/lib" ./configure
Followed by:
make -j 4

I'd installed libelf and libyaml with the prefix of /usr/local, hence the custom CFLAGS and LDFLAGS.

For the longer term, I'm guessing we'll need to closely track eco-systems such as Ruby, Python PHP and the like via Continuous Integration.

[Update:] Update the CFLAGS and LDFLAGS to have the correct content + explain what I did.

About SAN storage at Thoughtworks - Part 3

dynamicproxy — Tue, 01 Nov 2011 18:05:05 GMT

I'd earlier posted about how and why we selected an Oracle ZFS Storage SAN - the 7320. Read Part 1 and Part 2 for more context.

Having finalized that we want an Oracle ZFS Storage SAN, we placed an order for our SAN and eagerly waited for it. Some of our team received and installed it along with the local vendor, and then even put it to use. After a few days of use, we started to notice performance problems with our most important use case - VMs on VMWare.

Here's a checklist that you should followin:
a. Ensure that everyone understands what you're buying.
Modern day SAN devices include a lot of advanced technology that we usually don't see even on servers.

b. Get the support contract in place.
We had to seek support from Oracle once to understand why we were seeing poor performance, (more on that in the next post), and we were really pleased with the skills and competence of the Oracle Premium Support Engineer.

c. Validate the entire setup by Oracle.
Ensure that the SAN is installed and configured by Oracle themselves. Do not involve your local partner for this. You are paying money already, and Oracle wants to take the responsibility of the SAN. Let them do so. There may be commercials involved, so talk to your Oracle representative.

d. Ensure that the Welcome Kit and support are activated.

e. Ensure that the OS is upgraded.

f. Configure and test the Phone home service once.

g. Read the documentation at least once.
This is not your typical SAN, and this is not your typical filesystem. You need to know what you are doing incase you want to deviate from Oracle recommended configuration. If you want to configure custom disk pools, etc, remember that your various Windows and Linux and other SAN lessons _do not_ apply here. This is especially, especially important because the GUI makes things easy to configure and mis-configure, and if you're used to LVMs and what other SANs, you can easily make a mistake.

h. Run the SAN for a week before declaring it ready for production.
This is another important point. The SAN itself can actually work very well right from day one. However, you need to get used to the SAN since there's a lot of power packed in. There's also the GUI which you should get familiar with. If you need to reconfigure the SAN (a likely scenario), you would have the opportunity to do so.

Now that we have the above out of the way, let's move on to Part 4 - the SAN GUI !

Pluribus and upcoming network innovations

dynamicproxy — Sat, 29 Oct 2011 18:29:22 GMT

One company that I'm going to keep an eye out on is Pluribus Networks . They have the byline "Virtualization without Limits". Sunay Tripathi , one of the founders, is also of the driving forces behind the Solaris 10 TCP stack rewrite, and Crossbow, the network virtualization in the opensolaris kernel (and part of the upcoming Solaris 11).

Sunay recently blogged about "Network 2.0: Virtualization without limits", where he's written about a Network OS that controlls all switches, treating the network "exactly as one giant resource pool".

Pluribus are quiet about what they're doing, but Sunay has said that they "a network hypervisor that has semantics similar to a tight coupled cluster but controls a collection of switches and scales from one instance to hundred plus instances".

This is purely speculation on my part, but I do wonder if OpenVSwitch would be the management layer, or perhaps integrate/interoperate with Pluribus' NetVizor.

In any case, the future does look exciting :)

Interesting progress with Belenix development

dynamicproxy — Sat, 08 Oct 2011 11:57:24 GMT

During the past few weeks, I've been working a bit on Belenix as and when my illness permits.

So far, rpm, smart and createrepo have been working very well for me. I don't install from source, or download from openindiana repos any more. Everything that I need, I ensure that I build using spec files and rpmbuild, and install via a custom repository on my local computer.

I've been examining build systems (Jenkins/Hudson, Koji, Go), and am about to explore the Open Build System at OpenSuse. What I'd really like to find is something that lets the developer compose build pipelines, let us trigger builds on downstream components when an upstream component's build goes green, install build dependencies before triggering a build, and optionally, let users "promote" a particular package.

I realize that not all of the above would be directly possible, but I can't help feeling that surely someone somewhere would have wanted such mechanisms in place.

About SAN storage at Thoughtworks - Part 2

dynamicproxy — Sat, 08 Oct 2011 11:09:44 GMT

Given that the various high end SAN vendors all receive good reviews for their performance, let's see what some of these performance criteria are:
a. Excellent Disk IO !
b. Excellent Network IO !
c. Excellent caching mechanisms !
d. Optimized network stacks for various protocols !
e. Fancy reporting for the management !

There are also other interesting features such as:
a. being able to use multiple connectivity mediums (1G and 10G Ethernet, Fibre Channel).
b. Phone home, where the the device diagnoses issues if any, and sends the SAN company a message so that they can take pre-emptive action.
c.

Here are something that we liked about the Oracle SAN (the others have some of these, but not all).
1. A full blown Enterprise grade Operating System.
This SAN runs a version of OpenSolaris, based on the same enterprise grade OS that powers many of the world's performance critical environments. As screenshots in further blog posts will show, this SAN gives great performance + analytics while using neglibile amounts of CPU.

2. An excellent network stack.
Unlike some earlier versions of Solaris which were nicknamed "Slowlaris", Solaris 10 receive a TCP/IP stack rewrite. This rewrite was nicknamed "Fire Engine". Later, the developers went on to improve network performance in many other ways too, with quite some of those benefits making their way into this SAN.

3. The ZFS file system.
This filesystem was designed with some good thinking about and questioning of lessons learned in the past, and whether they need to be applied today or not. ZFS has long been acknowledged as being a really superlative filesystem, with ports to some other operating systems as well. There are efforts to write equivalents for other platforms such as Linux to which ZFS cannot be ported to for legal (licensing) reasons.

Some interesting features here are the modified elevator-seek mechanism, near-platter-speed access rates, end-to-end checksumming to provide you with greater reliability.

4. Caching.
One interesting benefit of having ZFS around is the improved caching. ZFS lets you specify read cache devices and write cache devices. These cache devices can be ordinary disks, but practically, everyone uses SSD devices for read and write caching. This means, you can start off as we did with having just one cache, and then using the Analytics to "size your requirements". In terms of storage, this means you could now use Analytics to determine whether you have more read or write operations, and whether you need a read or a write cache, how much, etc.

Sinze ZFS is part of the kernel, it can play with unused RAM and "soft allocate" some RAM for use as a read cache for very frequently accessed blocks of data (in case of LUNs) or file blocks or even entire files (in case of NFS and CIFS). What this means is, most of your frequently accessed data will reside in RAM and be served from there. In case the kernel needs to allocate some memory, it'll take away some from ZFS' RAM cache on a need basis.

So a RAM cache coupled with a Read and/or Write cache depending upon your requirements, can do wonders for your performance, over and above what ZFS acces speeds themselves do.

5. End to end checksums in the file system
This one deserves a point by itself. One problem that had hit me twice with the the MD3000i SANs, is that my applications complained about data errors. This was really dangerous for us at work, since this happened with active source code once, and with a VM in another case. The SAN reported faithfully that it had no corruptions in data, and yet I could see with my own eyes that I was in a bit of a mess with having to restore from backups, and rebuild from individual commits made during the day.

ZFS has this notion of an end to end checksum, where a block of data is checksummed, and then sent to the storage sub-system for a write, and there's a reference block that contains its checksum, with the grand parent block containing that parent block's checksum, etc. The intent is to be able to check whether a block that's been retrieved has been retrieved generates the same checksum or not. In case there's a checksum mismatch - say due to media error - ZFS knows to retrieve an identical block and help you get back your data. There's some documentation + diagrams that explain this better than I have.

Now, end to end checksums in a SAN do not guarantee to you at all that your server (the SAN's consumer/customer) will always get the data that it had sent to the SAN. There are many locations where the data could have got corrupted on the way to/from the SAN - the server's RAM (this is why you need ECC!), the network device driver or the OS' TCP stack may have some bug, the NIC card may be faulty, the network cable and switch may have some problems of their own, etc. You get the point.

But what end to end checksumming _will_ assure you about is, once the data reaches the Oracle SAN's network driver, from that point on till it gets written to the storage medium, it'll receive high fidelity checksum calculation, and this will be used to validate the data just before it's dispatched to the SAN's NIC.

For ideal quality, you should run ZFS on your server and have ECC RAM, but for those of us who have other use cases like having to run VMWare, you can at least rest assured that once your data is written to that pool, you can is even most of the worst cases get some or all of it back.

6. Excellent reporting.
You need to see this for yourself lest you consider me biased. Most other SAN devices provide you with what I call "defensive reporting", where the Storage Administrator gets to show that his SAN's giving great disk I/O operations. If someone were to ask a non-Oracle iSCSI SAN user "please tell me which of my VMWare servers is accessing which of the SAN's LUNs", then the storage admin would likely thrust some more defensive reporting in their face and ask them to get lost. If pushed to the wall, he'll simply call the SAN vendor's tech support team, and they'll ofcourse come scrambling to his aid to throw more jargon.

Apart from the fact that such an attitude doesn't take anyone anywhere, is the fact that better/alternate reporting simply doesn't exist on these other SANs.

With the Oracle SAN, you'll find that you'll be able to design and sketch interactive and drill down reports on demand. Drill down reports are awesome. Here's an everyday scenario "Hi, the VMs are slow, could you tell us what's wrong ?", "Sure... hmm... VMWare's CPU and RAM utilization continues to remain low, let me check the SAN" (By now, I trust the Oracle SAN's reporting since it' way more superlative). "Ok, I see high write operations from two VMWare servers, they're acessing LUN_EnvironmentD and Lun_EnvironmentJ, what're you guys doing on that ?" "Aah, there's a deploy going on" "Well, I see that the storage is well within the IOPS threshold" "Must be those script changes that we put in. Anyway, thanks for helping us arrive at this so soon". Apart from the time required to log on to the VMWare Management console and the SAN's own Web console, all SAN analytics literally takes as much time as it'd take to speak this conversation I've listed.

7. Direct support from Oracle.
Apparently, before Oracle acquired Sun, support would be available via channel partners too. Post the acquisition, Oracle now handles all support cases directly. There are pros and cons to this, I feel. There may be channel partners and their team members who may have wanted a career on configuring SAN storage, while continuing to play the role of a generalist. Now, they need to be lucky to get into a technologically diverse company like mine, or join Oracle !

On the other hand, since Oracle have their reputation on the line, you get to speak to people who have access to all manner of skillsets within the company. When crisis strikes, and your customers are screaming seeking escalation, you can now reply that this is the highest you can escalate. (see below).

8. Phone home.
Since I've seen this on NetApp and EMC too, I presume that all higher end storage vendors have this feature today. I know that even within a company such as Thoughtworks, not everyone is or will be as familiar with ZFS and related topics as I am. So, higher end SAN devices today run a number of self-diagnostics, and in case they find out any errors, they send some diagnostic data back to the SAN support team. Such teams take a judgement on what needs to be done, get in touch with the customer and set fixes in motion. These could be pre-emptively replacing disks, asking the customer to add more cache, recommend a reconfiguration, etc.

Alright, more in another blog post !

About SAN storage at Thoughtworks

dynamicproxy — Sat, 08 Oct 2011 11:08:16 GMT

At Thoughtworks, we use virtualization extensively, since a little before "cloud" became a buzzword. Until a few years ago, our platform of choice for Virtualization was VMWare. Since then, we also started to use Xen and KVM. We're yet to investigate HyperV in production use. As a software development company, we use VMWare in standalone mode, where our build agents are spread across a large number of servers (around 80 per server), and our UAT environments are made highly available on a number of VMWare clusters.

One of the requirements of a VMWare cluster, and that the VMs which need to be made highly available, should reside on a common storage device, typically a SAN . VMWare provide their own filesystem called VMFS, which is a distributed filesystem. If a SAN device presents a raw data store called a Lun, then VMWare accesses that Lun using the iSCSI protocol or via FibreChannel. If the SAN/NAS device can expose it's storage space via the NFS protocol, then VMWare uses that storage directly over TCP/IP. We initially purchased a Dell MD 3000i SAN device which played a useful role when we were operating at a small scale of around 3 TB. As we increased the number of VMs, the disk space required also grew. We finally stopped needing more storage for a specific project at around 5 TB of utilization (8 TB of usable capacity).

Given the larger number of VMs, and the number of parallel deployments of various applications by our automated deployment scripts (we made everything part of our Go Grid), we started to see performance issues. Another long pending problem was that of being unable to answer why our SAN was sometimes slow even though there was no deployment in progress at all (never mind parallel deployments). For e.g., if we knew that there were six environments running 2 - 3 different builds, we'd not know which particular build caused an increase in disk IO. The VMWare GUI tool gives you only so much data. Also, devices like the MD3000i do not have any such thing as iSCSI analytics.

Around this time, I also wanted to experiment with creating VMs within seconds.

Since I've been closely associated with Belenix and OpenSolaris technologies for years, I decided to give Solaris 10 a spin as a storage box for some time. We used ZFS and it's snapshot feature to set up VM collections in minutes. A typical VMCollection would comprise of a Domain Controller, some IIS servers, an Exchange Server, and some VMs running Outlook 2007. We also got amazing performance. All this on a box with just three disks of 1 TB each configured in what is known in ZFS parlance as RAIDZ. RAIDZ is similar to but better than RAID5, because it avoids the RAID5 write hole problem. Our VMWare servers would access our various ZFS filesystems - some over iSCSI and the others as NFS storage.

So, now that we understood that ZFS could for certain give good performance, we needed to solve the additional problem of identifying which VMWare server was sending how much disk IO to which SAN LUN. Enter DTrace. With the help of a close friend from Sun, we put together some dtrace scripts, and these provided answers to some extent. Why only to some extent ? That's because we often didn't know what exactly to ask DTrace to monitor.

Some months later, we decided to replace our MD3000i with a higher end SAN. Now I'd heard of The famed DTrace Analytics that somes as part of the Sun Fishworks product line. Today, this is called their ZFS Storage Appliance Line. However, since we were going to sink in a lot of money into a SAN, I wanted to be cautious and check out other popular vendors too.

Some colleagues and I got together and attended review sessions at NetApp and at EMC. Both NetApp and EMC offered more than what we'd imagined as part of their standard offering, and this was good to know. Unfortunately, at that time (mid 2010), NetApp did not have any analytics around iSCSI sessions. At the EMC review session at their Bangalore office, we saw a good demo and were pleased with their product overall. But the key feature which is very important for us - iSCSI session analytics - was missing. They did some some raw iSCSI analytics, but I learned that activating this monitoring locks up the controller in processing information, and VMWare servers and VMs get disconnected. This was a clear no.

After some discussions with Oracle Bangalore, we finalized an order for a 7320 Storage box, and life hasn't been the same again !

More in part 2 of this blog post.

Belenix - what's up ahead

dynamicproxy — Mon, 15 Aug 2011 17:14:04 GMT

Until a few months ago, the Belenix team had stopped all work on Belenix. We've been having a very good time at our respective day jobs, and that's been a very good trip so far.

Sometime ago, I decided to take a dive into Belenix again.

As of today, I've got a working rpm5 and smart package manager setup on my computer.

I've posted a roadmap here : http://mail.opensolaris.org/pipermail/belenix-discuss/2011-July/001510.html

I can report that I'm on track, and hope to have an IPS repository for rpm, as well as the beginnings of an rpm repository by the end of this month.

The recent news about Joyent's port of KVM to Illumos, as well as today's announcement of SmartOS are both very exciting. It looks like there's much to achieve ahead !

If it compiles and links, is it necessarily stable ?

dynamicproxy — Sun, 20 Feb 2011 15:42:49 GMT

Having worked on packaging and testing Belenix, I know that "if it compiles and links, it's not necessarily stable".

u_int8_t vs uint8_t

dynamicproxy — Sun, 08 Aug 2010 21:16:29 GMT

I consider this a good fix to have in place

http://dev.zuckschwerdt.org/openobex/ticket/15

C99 conformance is a nice thing to have in place

Fun times again

dynamicproxy — Sun, 08 Aug 2010 21:13:34 GMT

What a great technology-filled weekend.

Status:
Solve the following:
ld: fatal: library -lssp_nonshared: not found

Three new features in Fedora 14

dynamicproxy — Sun, 01 Aug 2010 14:48:27 GMT

http://fedoranext.wordpress.com/2010/06/09/fedora-14-three-new-features/

Fedora will use LZMA compression for the LiveCD at least.

rpm, dpkg, apt-get, yum, smart ?

dynamicproxy — Wed, 21 Jul 2010 18:41:00 GMT

One of the many attractions of a distro, is the quality of packages, and the ease of use of the package manager.

Choosing a package manager is easy: Use either yum/apt and put smart on top of it.

yum means the underlying package format should be rpm, while apt means the underlying package format should be dpkg.

Both rpm and dpkg today are good enough in their own right.

package formats also usually imply that the build recipes should be based on existing recipes which produce output in those formats (e.g. Fedora/CentOS for rpm packages, Ubuntu/Debian for dpkg packages).

The Belenix repositories at present have recipes in the form of spec files - these are in a format different from the Debian world. The package build recipes too are borrowed from SFE/Fedora. At present, the package quality at Fedora/CentOS is definitely much higher than that at Ubuntu. Debian packages are definitely of high quality, but one concern is that the packages are not updated as regularly as we'd like them to be.

Another concern is whether the Debian community would welcome contributions (such as Nexenta's enhancements to apt to support useful Solaris notions). This is a dilemma that can be easily resolved by talking to Debian, I think, even though they have in the past lashed out at well-meaning posts by Nexenta folk. This blog post (http://ianmurdock.com/solaris/no-good-deed-goes-unpunished/) makes things a bit confusing, though.

Update: Removed the Music label. The opensolaris planet was picking up the music label as the title of the post !

About the OGB decision

dynamicproxy — Wed, 21 Jul 2010 18:30:34 GMT

I decided to wait for a few days before blogging about my thoughts on the OGB's decision to ask Oracle to either appoint a liaison by August 16, or to resign on August 23 and had the community reins to Oracle.

While it is easy to accuse the OGB of being cowards, there are some points that we should remember:
- The opensolaris.org constitution mandates that there must be an Oracle representative (my wording, but that's the gist).
- The OGB members have approached Oracle via formal and via informal channels seeking responses from Oracle on how they intend to interact with the community.
- Till date, there have been informal responses to the OGB members, and two formal mails to various mailing lists attempting to provide reassurance to the community. There has also been a series of conference calls (which I have been unable to attend due to my own workload - more on that someday).
- The OGB members have to ensure that they do what's expected from them in their capacity as OGB members. Asking Oracle to honor the constitution and to respond to community questions are part of those expectations.
- One could expose the fact that all development of the OpenSolaris distro (different from the opensolaris community) is happening (if at all) behind closed doors by asking the distro community for a progress report. However, given that Oracle employees are stating that they cannot respond to anything they are not authorized to respond to, is already a signal in itself. However, Alan Coopersmith has told various community members that there are some critical bugs that are being worked on and that the distro would indeed be out soon. This was the message some months ago - I don't know the status today.

Given that there has been zero change in the above status for the past few months, I think the OGB did the right thing.

Also:
- Some of the OGB members are technically strong enough to have distros or package ecosystems of their own (Joerg with Schillix, Moinak with Belenix, Dennis with Blastwave).
- The others on the OGB have a proven track record of understanding corporates and open source (John Plocher and Simon Phipps).

It's not very clear what Oracle wants to do with the opensolaris community. It's not clear if Oracle understands how to deal with a community that thinks for itself (as against various communities of users who are interested in Oracle showcasing them new products).

I sure hope that Oracle doesn't decide to remain quiet and thereby damage the community. I also hope that the ON stays open and that new technologies are put out. After all, Linux awareness is more than AIX or HP-UX awareness because one is free and open, while the other two are not. We'd all want Solaris to rank high in awareness given the amazing technologies that it has.

Update: Removed the Music label. The opensolaris planet was picking up the music label as the title of the post !

What has Ubuntu done right ?

dynamicproxy — Wed, 21 Jul 2010 18:10:55 GMT

Like many other sysadmins, I'd rather have Solaris or Redhat/Centos as my server OS of choice than Ubuntu. Debian is great too, but I'm used to the Redhat way of configuration.

I was once a big fan of Ubuntu. Then I discovered Gentoo, and ended up learning a bit about setting up a desktop environment from scratch.

Despite rumours of a fall in package quality and of a decline in testing, people continue to use Ubuntu. I thought of jotting down some points here on why I feel this may be so:

1. bash command not found handler
See:
http://www.workswithu.com/2009/08/17/enhanced-command-not-found-hook-in-ubuntu-910/

2. Good desktop integration
This is almost as good as what OSX or Windows 7 have today by way of MIME type handling, file associations, etc.

3. Good collection of updated packages
The wide variety of packages help users understand and have confidence that they can do a wide variety of things with Ubuntu.
Other distributions (Linux, BSD-style) too have a variety of packages, but the public perception is that the Ubuntu package repositories are very much up to date, and that they have a wide variety of packages.

There are some other points that set Ubuntu apart. See

http://www.workswithu.com/2009/04/23/four-simple-features-that-set-ubuntu-apart/

Some of these points can be fixed on Belenix and on other distros using technology, while others need a lot of perception management.

Update: Removed the Music label. The opensolaris planet was picking up the music label as the title of the post !

Belenix - the next steps

dynamicproxy — Wed, 14 Apr 2010 20:20:35 GMT

Given the uncertainity around the OpenSolaris 2010.03 release and the absence of any clear information from Oracle on the opensolaris code base, the OpenSolaris distro, and the opensolaris community, it is but natural to wonder as to what would happen next.

Moinak and I have been discussing these topics ourselves nearly everyday.

Our thoughts are as follows:
- Let the OGB meet together, and perhaps also meet a senior Oracle person such as Dan Roberts, and understand what Oracle's plans are as well as the reason for the silence.
The OGB represents the community, and is supposed to represent and oversee the Community Groups, so it's best the OGB take optimistic and friendly steps toward Oracle.

- It is not yet time to fork the code base. Let us learn conclusively that Oracle will not be contributing code, and then take a decision along with the opensolaris community.

Our plans at present are to get the Belenix 0.8 release out, and to then work with the community and the OGB to talk to Oracle and decide on further steps.

See you at the Sun Tech Days !

dynamicproxy — Sat, 20 Mar 2010 16:53:31 GMT

I'm looking forward to giving my talk at the Sun Tech Days the coming week.

Rolling Solaris into the organization has been an interesting adventure. Buying a Fishworks unit was not working out, though I tried to make that happen twice. So, onto Plan B.

Plan B:
- Set up a cheap storage box using Solaris 10.
- Export NFS mount points and backup build artifacts onto that.
- Export iSCSI LUNs and use for VMWare
- Coach fellow sysadmins every step of the way, and make them manage these boxes
- Show how fast setting up VM eco systems can be when you use ZFS snapshots and clones (e.g. six VM environments ready for configuration in < 5 mins).
- Move on to Zones :)

Confidence building was the biggest task. I was helped by a few disks failing, and my colleagues discovering how awesome ZFS repairs can be.

We also use Linux DM based mirroring (since LVM mirroring is not upto the mark). ZFS mirroring speeds are easy to convince anyone, when they see even creating an empty 1 TB ext3 DM mirror can take upto six hours.

A very important thing that I've learned is that evangelizing Solaris technologies to the average Indian sysadmin requires a lot of Indianization, where we speak of Indian realities (dealing with people, management, budgets).

I'm looking forward to my talk on all of this at the Sun Tech Days on Thursday.

This is the last Sun Tech Days in India. Later, perhaps, it'll be part of the Javaone in India, perhaps as part of the "Develop" event.

Belenix 0.8 Alpha - with KDE 4.2.4

dynamicproxy — Fri, 28 Aug 2009 15:05:46 GMT

Here's a snapshot of my desktop which I use at work
.

That's Belenix 0.8 Alpha running KDE 4.2.4 compiled with GCC 4.4. I backed up my data, installed OpenSolaris 2009.06, and then ran the install_belenix script after modifying it to point to my local Belenix mirror. Install took 20 minutes, and I now have OpenSolaris as well as Belenix 0.8 Alpha in my GRUB boot menu.

I've taken Wine from Triskelios' builds since we need to patch GCC 4.4.1 in order to get it to compile Wine on Belenix. This is good enough to run Lotus Notes 7.0.1. I do face two errors at startup related to z:\, which I've not got around fo fixing yet.

Also, I've taken rdekstop from www.sunfreeware.com since I needed to get rdesktop running first and didn't have time to set up a GCC 4.4 based dev environment (which I now have).

Some bugs that I've noticed so far:
- The KDE Lock tool doesn't let me unlock ! This is a critical bug which need some investigation.
- The cursor in Konsole is offset by a fre characters from the point of actual character entry.
- KWin sometimes thinks that the Alt key is continuously pressed. I "solve" this by either restarting KDE (Control Alt Backspace), or I SSH into my box and kill the startkde4 process. I've had to do this four times during the past week.

We have core dumps for the above and are considering either investigating further, or simply moving to KDE 4.3

Other bugs that I've noticed include:
- Okular doesn't support PDFs due to some issues with how poppler has been built.
- On a colleague's laptop, the CPU fan started to run at full speed. This is the first time that I've notived this issue.

There may be other issues too, but I've not made time to test them yet.

We've ofcourse had a number of bugs with the install_belenix script, each of which has been fixed.

We welcome you to try out Belenix 0.8 Alpha and report issues. Even if you have OpenSolaris already installed, you can go ahead and install Belenix since it'll neither damage your existing data, nor format anything. The script will ask you for a user name, and will create a separate user account for you.

wget http://www.belenix.org/binfiles/install_belenix
chmod +x install_belenix
pfexec ./install_belenix

Let's make the web fast

dynamicproxy — Mon, 03 Aug 2009 10:42:29 GMT

http://code.google.com/speed/articles/

What many may not know, is that having great disk i/o and a good file system can help in certain situations.

So also with an excellent network stack.

Three of Four JRuby core team move from Sun to EngineYard

dynamicproxy — Sun, 02 Aug 2009 10:33:13 GMT

http://www.engineyard.com/blog/2009/jrubys-future-at-engine-yard/

Non-IE browser support -> Important for non-IE users

dynamicproxy — Sat, 25 Jul 2009 13:11:41 GMT

As a sysadmin who doesn't use Windows at all, I have to RDP over to a Windows server, start IE on that computer, and administer switches from there.

There are also sites like www.cisco.com which are not Firefox friendly at all.

I recently wrote to the marketing team at Promise (the SATA Controller company), pointing out that their website is not usable by non-IE users. They wrote back stating that the website will be revamped soon.

At my company, the marketing team tests all functionality to ensure that mose browsers that we can think of work fine. This includes IE 6/7/8, Safari (OSX/Windows), Firefox (OSX/Windows/Belenix). We don't test with Konqueror or with Webkit yet, though I think those two should be included too.

As a developer, I know that web standards compliance requires a bit of diligence but that this can be achieved.

Quote -> "ZFS is the most amazing technology I've seen in recent times"

dynamicproxy — Mon, 18 May 2009 04:40:03 GMT

On Friday night at work, some of us were discussing our plans for the weekend. I mentioned that I'd visit I2IT - a college in Pune - and give a presentation on OpenSolaris and ZFS to the students. Ever on the watch for a chance to showcase ZFS, I gave my colleagues a quick demo on ZFS.

We had a look at creating a snapshot, deleting files, and then recovering files from the .zfs folder as well as by rolling back to the snapshot. I also cited some disk performance numbers (5 mins for a full SVN checkout of a particular project on my laptop, vs 30+ mins for others who use Windows on the same laptop model). I also mentioned the notion of pools, of how one can transparently add storage, etc.

I will be showcasing ZFS sometime this week to the entire office.

While we were leaving, my colleague remarked "ZFS is the most amazing technology I've seen in recent times".

Reflecting on it a bit, I couldn't agree more !

Some links on Cloud Computing

dynamicproxy — Sun, 17 May 2009 05:22:57 GMT

Here is a useful link that helps explain Cloud Computing : http://industry.bnet.com/technology/10001746/fog-is-lifting-on-cloud-computing/

The Open Cloud Consortium has as list of software related to Cloud Computing. There may be others too.

According to the statements on Sector listed at that page, Sector is supposed to be twice as fast as Hadoop .

Though I consider myself too old to understand the medium, I'm intrigued by the effort needed to keep Facebook running. Facebook's cloud computing related software, called Thrift , is available for download too.

Matt Asay asks Cloud computing: A natural conclusion of opensource ? . This is an informative read, because Matt provides perspectives on how users are no longer interested in the underlying technology, but in how they access that technology, and how inter-operable the data is.

Demos, Software, and getting them to an audience.

dynamicproxy — Sun, 17 May 2009 05:05:40 GMT

My roommate is excited about opensolaris. Being a .NET developer, he wished to help create a mono package for Belenix. While I walked him through an exercise on building mono from source, I realized that we had a "broadband" line which gives us just 20KBps. While trying to get mono to build (we've not done this yet due to a corlib.dll version mismatch with mcs), we had to wait a while for mono and for monolite to get downloaded.

Yesterday, I'd presented on Belenix and ZFS at a local college I2IT, and had fallen short on DTrace and Zones demos. My custom Belenix installation has a broken zones config, and I've not made time to learn to write DTrace scripts.

I also learned that it is impossible for the students of I2IT to download the Belenix ISO, or any of the tools and software that one could run on. Though they have a Sun Campus Ambassador who is doing a good job, he can only distribute what he receives as part of the ambassador program. Anything more, and the students have to fend for themselves.

All opensolaris evangelists should have a good collection of demos for various topics, familiarity with the various topics that they are going to demo, CDs/DVDs to give away, and learning material (or at least a printout of URLs that people should visit - in case you run out of CDs/DVDs).

Expecting people to download content is unrealistic - they may not have the bandwidth, and the sheer chore of having to download something could be a deterrent. This is a reality in India, and when we evangelize, we need to have various tools ready and available on distribution media.

It may be worth having a demo CD all by itself, in case the basic Belenix CD does not have space. An alternative to consider would be having DVDs.

Here are some demos that I think we should have, along with the requisite software where possible.
- Entertainment, Games, Graphics tools, OpenOffice
- Communication Tools (IRC, Browsers, streaming-content servers and clients, file download tools, Skype in a Linux Zone, Jabber, etc). See note below on Skype.
- Web app stacks
- Wine for running Windows apps, along with some free Windows apps
- Development Tools
- Using opensolaris tools to solve college assignments, and to understand subjects better.
We could have sample code that illustrates a topic, and DTrace scripts that helps the programmer see how the code behaves on a running system. This would be a very useful demo, imo.

A lot of the above is already available thanks to the efforts of the opensolaris community, the various blogs, and the documentation team. The next round of work would be to collate all this together, improve where necessary, and to create DVDs. There'd be a lot of testing too.

The creators/communities of the various tools would be a good group to source demos from.

Along with all this, we should also have instructions on setting up a local mirror, a DVD of the Belenix repository which could be used as the seed for the local mirror.

Note on Skype redistribution: We're allowed to re-distribute Skype, provided we send them an email, and comply with various conditions there ( a bit of a task, but can be achieved).
http://www.skype.com/intl/en/legal/promote/distribute/
http://www.skype.com/intl/en/legal/promote/materials/
Would it be legal to setup Skype in a zone and distribute that ? We should find out.

Note on Sun Studio Redistribution:
We need to have an answer once and for all - Can the excellent Sun Studio Compilers be redistributed by non-Sun distros ? We've initiated conversations by have not followed up on them.

Opensolaris/ZFS session at I2IT - musings

dynamicproxy — Sun, 17 May 2009 04:08:44 GMT

Amey and I presented on opensolaris at Pune yesterday. Amey has put up a report here

Here's my own analysic.

What went well:
- Rapport with the audience

- ZFS demos
In my experience with showcasing opensolaris, ZFS alone sometimes convinces people that opensolaris should be considered seriously.
To Sun's ZFS team: I hope to do you proud with even better ZFS demos !

What didn't go well:
- Some students "forgot" that I'd mentioned that Belenix was a first class KDE 3.5.x environment.
They wanted to know at the end how Belenix compared to a KDE environment. I blame myself for such questions, since I should have established beyond doubt that Belenix has just about everything that a KDE environment has to offer.

- I forgot to prepare and take along Belenix CDs :(
I was simply flooded with work, and am anyway guilty of not managing my time well.
Now that I'm used to a 10 Mbps line at work where ISOs are downloaded within an hour or so, and because I know that I2IT is a premier institution, I assumed that the students would be able to download the Belenix ISO themselves. They told me that this was not possible given that they had limited bandwidth at college and that they had to go through a committee for anything over 30 MB in size. This excludes even GParted, btw, which is over 30 MB in size.
This made me realize that if I get used to having high-speed bandwith despite being an Indian staying in India, it is unrealistic to expect the IPS team to understand that the world cannot download lots of content all the time.

Future Directions:
- Emphasize how Belenix is a first class KDE environment.
Talk to the KDE evangelizing team for ideas here.
Mention how the next version of Belenix could set the gound for opensolaris being _the_ KDE platform of choice.

- Different presentation styles for students vs working professionals.

- Demos, demos, demos
This needs a separate blog post by itself.

- Have CDs and the opensolaris Learning Guides ready for use.

Bottom line: Be prepared.

Dependency Injection or Getters and Setters ?

dynamicproxy — Sun, 26 Apr 2009 14:03:43 GMT

One blog post that I read just now is "Complexities of APIs" where Ketan observes how a service locator gets in his way.