November 29, 2009

UTF8 collation support for Croatian, Bosnian and Serbian (latin) in MySQL

There has been a long lasting problem of collation in MySQL for Croatian language - it was impossible. When, at that time, Yugoslavian keyboard layout was invented, it was designed to cover all the languages from all republics. It covered all Slovenian characters (plus couple of characters that they don’t have), but not all Croatian (it missed ‘nj’, ‘lj’ and ‘dž’). When Yugoslavia fall apart, all the republics just took already wide spread Yugoslav layout. For Slovenian layout, that was great, with exception of including characters they didn’t have. For Croatian and Serbian latin, well, not that great…

You see, now we type letter ‘nj’ as a combination of ‘n’ and ‘j’. Same thing with ‘lj’ and ‘dž’. That wouldn’t be that bad if every word, containing ‘l’ or ‘n’ and ‘j’ together would be pronounced as ‘lj’ or ‘nj’. For example, we have two words that we write the same (injekcija), but pronounce different. In one case with say it with ‘nj’, and in the other as ‘n’ and ‘j’. Talking about the bad choice of deciding to use ‘nj’ as a character for that letter. There are also examples of ‘dž’.

As you can see, until we put ‘lj’ and ‘nj’ characters on the keyboard, we will never have correct sorting in any database. Good news is that those characters exists in Unicode and that’s why we have ‘hr unicode’ layout in Xorg. Too bad nobody uses it.

Until that’s sorted out, I’m happy to announce that MySQL just accepted a patch that makes possible contracting non-ascii characters, meaning that we can now have sorting rules for ‘dž’ (more about that at http://www.collation-charts.org/articles/croatian.htm). As a result, utf8_croatian_ci and ucs2_croatian_ci collations were created and added to MySQL 6. Since Alexander Barkov was so kind and provided a patch for MySQL 5.1, I’ve created packages for Ubuntu. I’ve also modified that patch so that it works with MySQL 5.0. If you need this feature, go add my PPA to your sources.list:

https://edge.launchpad.net/~ivoks/+archive/mysql-hr/

It’s important to realize that this patch contains very intrusive change in collation mechanism, so it’s not just a patch for Croatian collation. People from Bosnia and Herzegovina, Monte Negro and Serbia (latin) can also use this collation for their languages. It does not cover all problems (’injekcija’ and ‘injekcija’ for example), but at least words starting with dž won’t be at the end of the sort :)

November 26, 2009

Server Team 20091125 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION from previous meeting

ACTION: ttx to review status of bugs 455625, 460085 and 461156 for any missing info ACTION: mathiaz to compile a list of easy merges for publication

Check blueprint status and progress for the week (mdz)

mdz reminded that the list of blueprints is used to track plans for Lucid. The focus is now on drafting the specifications after last week discussions at UDS. Work items should also be added to the blueprint whiteboard so that a burn down chart can be generated during the cycle. Once the wiki page is written and the work items have been defined the status of the blueprint should be set to Review.

All of the blueprints should be ready for review first thing Monday morning.

Assigned and to-be-assigned bugs: http://qa.ubuntu.com/reports/team-assigned/canonical-server-assigned-bug-tasks.html (mdz)

Nothing was assigned to the team. Most of the bugs seemed to be SRU-related.

Work items tracking (mdz)

 

mathiaz asked how to handle work items that can’t be defined up-front as they depends on completion of existing work items. mdz suggested to create work items for each of the proposed changes. If some of them can be skipped, it’s easy to skip them later, but we don’t want to forget any. The most important thing is that the list is at approximately the right level of granularity, so that we make steady progress through the list. Work items need to fit into a 1-2 day chunk of work.

Weekly SRU review (mathiaz)

 

Only the hardy nomination list had one bug to be reviewed. The last two weeks of fixed bugs have also been reviewed for potential SRUs.

Spamassassin update

 

ScottK asked about the status of Spamassassin in Lucid. mathiaz replied that Daviey had been investigating the situation with upstream. He also suggested to define work items in the associated blueprint even if the drafter doesn’t plan to do the work. Documenting what needs to be done may help in getting things moved forward by other people.

Agree on next meeting date and time

 

kirkland to discuss a new time slot with maria.

Next meeting will be on Wednesday, December 2nd at 14:00 UTC in #ubuntu-meeting.

November 23, 2009

I'm Running Ubuntu (2009-11-08)


I'm training for next year's Austin Marathon, on February 14, 2010. This will be my fourth marathon, having previously completed Motorola Austin/2004 (4:12:15), Freescale Austin/2006 (4:08:27), Marine Corps WashingtonDC/2007 (4:49:26 -- with ankle sprain). Since I didn't have a blog at the time, I never posted a race report. I did write one for each of those races. I'll post a retrospective on each of those races here. If you're interested, you might subscribe to my running tag, as I probably won't post those to Planet Ubuntu.

In preparation for that race, I ran Austin's Race for the Water 10 miler. I thought some of you might enjoy the custom tailored technical t-shirt, which proudly states:

i'm running ubuntu


The back of the shirt asks:

ARE YOU RUNNING UBUNTU?

If you're interested in the results, I completed the 10 miles in 01:28:40 (that's an 8:52/mi average). I was shooting for a 9:00/mi average, so clearly I was pleased with the result!

:-Dustin

November 16, 2009

UDS Lucid


This week, Dallas hosts the Ubuntu Developer Summit for the Lucid Lynx release. This is the key moment where we define what will be done for Ubuntu 10.04 LTS, and discuss how it will be done. There will be plenty of interesting sessions in all the tracks, and sometimes I wish I could attend two sessions at the same time. In the server track, Monday will have a special focus on applications, Tuesday on cloud images, Wednesday on Ubuntu Enterprise Cloud, Thursday on infrastructure sessions, and Friday on testing.

I’ll just highlight one session that I think represent the QA focus of this release: “Server usability papercuts”. The 100 papercuts project focused on common annoyances, easy-to-fix bugs that hinder the user experience in Ubuntu Desktop. It was a great idea, and I think a great success. Several members of the server team think this can be transposed to the Ubuntu Server product as well. Some server packages may be counter-intuitive, suffer from bad default configurations or force the systems administrator to stupid repetitive tasks. Some other packages might not work well together, while it is good practice to use them together. We could identify some of those and fix them, to improve the general server experience for this release. Interested ? Join us today, 15:00 Dallas time !

November 14, 2009

Results of the Ubuntu Virtualization Survey

A big thanks to everyone that participated in the Ubuntu Virtualization Survey. I am pleased to share the results with you now.
I will provide a few of my own observations, but we are very interested in your own conclusions!
  • There were a total of 354 responses -- excellent feedback!
  • Nearly 2/3 of all responders use virtualization on Ubuntu every day -- wow!
  • Over 3/4 of responders have VT acceleration -- that's overwhelming, I think, and it supports our focus on KVM.
  • Still, there's 21% of responders who cannot use KVM. kqemu has been deprecated by upstream QEMU, so I think VirtualBox represents the best option at this point for non-accelerated virtualization.
  • 36.7% of responders most use VirtualBox, 22.6% most use KVM. VirtualBox is in Universe and essentially unmaintained by Canonical (though some community individuals are doing an excellent job maintaining it!). I don't know what the business opportunity is around VirtualBox. But it is clear that it's popular among Ubuntu users. People really like the interface and the usability. And we could probably really improve the experience for a large number of Ubuntu virtualization users with some dedicated Canonical effort to clean up the VirtualBox bug backlog.
  • My survey design was evidently flawed on Question #3, as a large number of people "wrote in" an "EC2" answer there. This is an interesting approach, as it diminishes the importances of having VT on the local system.
  • In terms of interfaces, virt-manager and virsh are both lagging behind kvm-from-the-command-line and VirtualBox. I don't know if this means that we should, or should not invest more in the libvirt-based tools. Is the lack of a good GUI for KVM hindering its adoption? I think this data says so...
  • Finally, the overwhelming majority suggests that better documentation is simply required for Ubuntu virtualization. I wonder how we should approach solving this? Is this something that we as engineers should be able to just crank out ourselves? Or should we tap into the Ubuntu-Documentation-Team, and attempt to rally a virt-documentation blitz from some more skilled tech writers?
In summary, I think the most important observations are that:
  • The overwhelming number of respondents have access to VT hardware.
  • VirtualBox is quite popular in the wild, despite a lack of Canonical investment.
  • The lack of a better user-interface is hindering KVM's adoption.
  • Better documentation is undoubtedly requested.
Are there other observations you'd like to share or conclusions you can draw?

We will be in Dallas next week for the Lucid Ubuntu Developer Summit, discussing the future of Virtualization on Ubuntu. Thank you so much for your feedback!

Cheers,
:-Dustin

November 12, 2009

Introducing Testdrive!



I'm pleased to introduce a new package I have created for Ubuntu called testdrive!

Testdrive makes it simple to run any Ubuntu release in a virtual machine, safely, and without affecting your current Ubuntu installation.

This is a great way to "try out" the Ubuntu release beyond your current version, before upgrading. For example, if you're still running Ubuntu 9.04, you could testdrive Ubuntu 9.10 before committing to the upgrade.

You could also testdrive a different flavor of Ubuntu, such as Xubuntu, Kubuntu, Netbook Remix, or the Ubuntu Server. This is great way of learning more about the Ubuntu galaxy, as well as introducing yourself, to the wide world of virtualization in Ubuntu.

I expect that testdrive will be very useful to Ubuntu developers, testers, and bug triagers during the Lucid development cycle, as these people will be able to test Lucid's daily ISOs throughout the cycle, and in particular at the release milestones for ISO-acceptance-testing.

Prerequisites

Testdrive can use either KVM or VirtualBox to host the virtual machine. You should have either one of these installed on your system. If you're using KVM, you need to have at least kvm-84, which is available in hardy-backports, intrepid-backports, jaunty, and karmic.

You should also have enough disk space available in your home directory to store one or more ISOs, roughly a 1 GB or so.

Installing Testdrive

To install testdrive:

Running Testdrive

To run testdrive from the command line, you just need to provide the URL to an ISO that you want to test. This can be an http, ftp, rsync, or file style URL. The ISO itself will be cached in your ~/.cache/testdrive directory, such that subsequent runs will only need to perform incremental downloads.

From the command line you could do something like the following:

testdrive -u rsync://cdimage.ubuntu.com/cdimage/daily-live/current/karmic-desktop-i386.iso
testdrive -u http://mirrors.kernel.org/ubuntu-releases/9.10/ubuntu-9.10-server-amd64.iso

You can also add some other configuration details in your own ~/.testdriverc file. Simply copy /etc/testdriverc to ~/.testdriverc and edit as you like. Once you have done so, you can simply launch testdrive from the menu, with:
  • Applications -> System Tools -> Test Drive and Ubuntu ISO

Testdrive-GTK

Rick Spencer, Manager of the Ubuntu Desktop Team, has used quickly to draft a GTK front-end for testdrive. Hopefully, testdrive-gtk will make it into the archive for Lucid soon, and provide a nice, pointy/clicky way of choosing the Ubuntu release you'd like to testdrive.

UDS

I'm giving a plenary talk at the Ubuntu Developer Summit in Dallas, Texas next week, where I plan to demo testdrive, as one example of what we can do with KVM and Virtualization in Ubuntu. If you have been reluctant to try Ubuntu Virtualization, testdrive is a really easy way to get started!

:-Dustin

November 11, 2009

Server Team 20091111 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

ACTION: nurmi to help investigate/validate/fix bugs 455625, 460085 and 461156

  • Daviey to investigate spamassassin status in lucid: Done

  • mathiaz to update the fridge: Done

UDS: server track planning

mdz will schedule the server track before the end of the week. To make sure he doesn’t miss your blueprint, please name it server-lucid-* and set him as approver. There can be two parallel sessions in the server track (as with other tracks), so you should subscribe to the blueprints you are interested in to try to avoid scheduling conflicts.

Merging, merging, merging

Lucid is open for merging, which is a great way to learn packaging for starters. Since compiling a list of easy merges seemed to help last time, mathiaz will compile it again this time and blog about it.

ACTION: mathiaz to compile a list of easy merges for publication

Open Discussion

nijaba mentioned bug 450044, which allowed to spot that ubuntu-server was not a bug contact for euca2ools (now fixed). Most of the crew is expected to enjoy the great(?) texan winter in Dallas next week.

Agree on next meeting date and time

Next week is UDS Lucid, so next meeting will be on Wednesday, November 25th at 14:00 UTC in #ubuntu-meeting.

November 06, 2009

Server Team 20091104 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

ACTION: kirkland to add a recipe covering virsh to https://help.ubuntu.com/community/KVM/VirtManager

  • nurmi to investigate bug 455625

ACTION: nurmi to investigate bug 455625

UDS Lucid preparation

ttx created a wiki page for 10.04 input from server team. mathiaz reminded that there will two tracks dedicated to the Server team like the last UDS.

Assigned and to-be-assigned bugs

 

A discussion about adding python-software-properties to ubuntu-standard took place. soren noted that if the request was to add the package to the default server install adding it to ubuntu-standard may not be the best place. zul will look into that with the Fondation team.

ScottK mentioned that it looked like courier was pretty badly broken at the moment and in need of at least a serious triage effort.

Eucalyptus Karmic SRU

 

mathiaz announced that a first round of bug fixes for UEC in Karmic had been prepared. Packages are being uploaded to -proposed. Help in testing them will be welcome once they’re accepted by the SRU team.

Daily EC2 images

 

smoser reminded that ‘daily’ images of karmic for EC2 are now automatically published twice weekly. Image removal is now covered by the new policy In short, 5 builds of each release/arch are kept in ‘testing’ (daily builds). When a build ‘falls off’, it is made private for 30 days before deletion. This policy covers kernels and ramdisks in ‘-testing’ also. The code is available in LP – feedback is welcome!

spamassassin review

 

ScottK pointed to the state of spamassassin in Lucid: the upstream project hasn’t done a new release in over a year. Questions about its fate (leave it stale, switch to svn snapshot, other solution) should be discussed during UDS.

ACTION: Daviey to investigate spamassassin status in lucid

Agree on next meeting date and time

 

ACTION: mathiaz to update the fridge

Next meeting will be on Wednesday, November 11th at 14:00 UTC in #ubuntu-meeting.

November 05, 2009

Skills Matter presentation: What is Ubuntu cloud?

Thanks a lot to everyone that came to Skills Matter tonight for my presentation. I have really appreciated the quality of your questions and hope that my answers were satisfying.

read more

November 03, 2009

More Ubuntu Server Edition statistics

Some people may say that I am a statistics junky... Well that's certainly true! But what do you want, being a product manager for a product that does not require ANY form of user registration, you have a tendancy to cling to any piece of data you may find that shows that you are not working in vain. Indeed, and to the opposite of most, if not all, of our competitors, we have absolutely no way to determine what is our install base.  We don't control our mirors, we don't have any ping back home mechanisms, and we are not considering adding any.  So, here I am, collecting as much information I can from outside sources...

Well, october was not too bad in that sense:

read more

Register Bloodied by Lack of Research


Typically, I read and respect The Register. They usually run intriguing technology articles that make me think.

I'm quite disappointed with today's carelessly researched piece:
Specifically, these paragraphs regarding eCryptfs:

Encryption proved a hurdle for Ubuntu forum member XXXX, who decried the lack of automation on encrypting his home partition.

"I had chosen to encrypt the home partition when installing 9.04 and then wasn't able to get the passphrase command to complete the encryption process to work properly," XXXX wrote.

Finally, after a late night and getting some advice online, XXXX wrote: "I certainly wish the encryption mounting process was more automated like everything else is!!"

Lack of automation? In Ubuntu 9.10, encrypting your home directory is a matter of selecting a check box in the installer:


That's it. 9.04 Encrypted Home upgrading users simply run update-manager and upgrade all packages to 9.10. Their home directory encryption is not affected by this.

The author of this article found one post in the Ubuntu Forums poorly articulating an issue with home directory encryption and suddenly Ubuntu 9.10 users are getting "bloodied" by encryption in Ubuntu? Seriously?

The Register, we are expecting more from you...

:-Dustin

November 02, 2009

Ubuntu 9.10 Byobu and OpenWeek Session



I thought I would provide a brief set of highlights about Byobu accomplishments during the Karmic development cycle, now that we have released Ubuntu 9.10.
Also, I'd like to promote my Ubuntu Open Week Presentation on Byobu, which is scheduled for 18:00 UTC, tomorrow, Tuesday November 3, 2009. It will included a live demonstration, in Amazon EC2. Be prepared to join us in #ubuntu-classroom on irc.freenode.net, and SSH into a guest session and participate in the presentation and discussion!

Statistics
  • Renamed the project, from screen-profiles to byobu
  • Uploaded about 43 times
  • Fixed approximately 74 bugs
  • Contributions from 9 different people
Features

Performance
  • Tremendous performance improvements across the board -- all status scripts
  • Better timings for status script run frequency reducing overall load on system
  • Cache files in /var/run/screen ramdisk for better performance
  • Check updates immediately, and only when package lists are updated
Usability
  • Dynamically reload profiles on byobu-config changes
  • Interpret ESC as "cancel" in the byobu-config menu
  • Dynamic keybinding updating on changed escape sequence
  • Support dynamic enabling and disabling of keybinding sets
  • Better default window handling
  • Show the MOTD on first launch
  • Update ssh authentication socket on re-connection to an existing session
Appearance
  • Design improvements, per review with the Ubuntu Design Team
  • Enable 256 color support
  • Color coding, vim folding, and formatting the detailed status output
  • Improve colors of status notifications, use bold for numbers, non-bold for units
  • New distro logos
  • Set the window title appropriately
Compatibility
  • Improved support for non-bourne shells, busybox environments, non-x86 architectures, and non-Ubuntu operating systems
  • RPM packaging spec file
Accessibility
  • Comprehensive documentation
  • Internationalization
  • Use UTF8 where possible
Manageability
  • Make status monitoring more configurable (ie, watch a different ethernet interface or disk partition)
  • New status scripts (ip address, disk availability, cpu temperature, cpu fan speed, mail, reload required)
  • Dropped nasty /usr/bin/screen diversion
  • Created a byobu-janitor utility that collects all migration hacks to a single place, and runs on a profile refresh
Hope to hear from you tomorrow during the Open Week session!

:-Dustin

Ubuntu Karmic Release Party in Austin

30+ Ubuntu enthusiasts, free software developers, hackers, beer drinkers, and spouses attended Austin's Karmic Release Party on Thursday, October 29, 2009, celebrating the spectacular Ubuntu 9.10 release.

We filled half of the dining space at Aussie's, an Australian-themed volleyball beach bar--in honor of our Koala mascot and the Ubuntu Enterprise Cloud (UEC) powered by Eucalyptus.

I got there about 3 hours early, and setup a UEC instance in the corner, using:
  • Linksys 310N wireless router and gigabit switch, flashed with DD-WRT, wirelessly bridged to Aussie's free WiFi and the Internet
  • 1 Thinkpad X61 (dual 2.0GHz, 4GB, 250GB), Ubuntu 9.10 amd64, running the Eucalyptus Cloud/CC/SC/Walrus services; i.e., the cloud front end
  • 2 Dell Vostro (dual 2.4GHz, 4GB, 320GB), Ubuntu 9.10 amd64, running the Eucalyptus NC services; i.e., the cloud nodes
Over the course of the party (6pm - Midnight), I did roughly 4 demonstrations of UEC on the local installation, showing the web interface, command line EC2-compatible tools, running instances, deploying appliances, terminating instances, and answering a number of excellent questions from our party goers. I also brought a Watt-meter and demonstrated PowerNap -- the unique feature of UEC that enables it to be the most energy efficient private cloud deployment around. Oooh...aaaaah :-)

Unfortunately, I was too busy talking and doing demonstrations, so I didn't take any good pictures this time. Sorry!

:-Dustin

October 30, 2009

Solar Installation - Part 5

As of 11am today, my PV system generated its first Megawatt-Hour of electricity


Current date/time: 30-Oct-2009 16:00:09

Daily Energy = 27.766 KWh
Weekly Energy = 123.155 KWh
Monthly Energy = 634.408 KWh
Yearly Energy = 1025.502 KWh
Total Energy = 1025.683 KWh
Partial Energy = 935.313 KWh


Fri Oct 30 16:01:29 CDT 2009

Current date/time: 30-Oct-2009 16:01:25

Input 1 Voltage = 300.503571 V
Input 1 Current = 9.931437 A
Input 1 Power = 2984.432373 W

Input 2 Voltage = 293.223511 V
Input 2 Current = 4.005443 A
Input 2 Power = 1174.490112 W

Grid Voltage Reading = 244.775589 V
Grid Current Reading = 16.397514 A
Grid Power Reading = 3985.880371 W
Frequency Reading = 60.003841 Hz.

DC/AC Coversion Efficiency = 95.8 %
Inverter Temperature = 49.636932 C
Booster Temperature = 42.354176 C


I owe a big thanks to Curt Blank, who wrote aurora, a GPL program that communicates with my Aurora Solar Inverter and generates the statistics shown above. We have been talking over email through the last few weeks, trying to debug a few issues with my inverter, which is a newer model than the one he developed his software against. I have a cronjob that polls my inverter every 15 minutes, logging statistics about my solar power generation. And I will be packaging aurora for Ubuntu Lucid shortly. Thanks for all your help, Curt!

The system has been running now for almost 50 days. I have averaged about 20.6 KW-hours of generated energy per day over the last month-and-a-half, with a maximum of 36.2 KW-hours in a given day. Now the days are getting shorter in the Northern Hemisphere, and we're in the rainy season here in Austin. So I'm looking forward to the long, sunny days next summer.

My PV system saved about $80 on my October 2009 electricity bill, which is really close to my predicted target.

Stay tuned for one more installment in this series, regarding the actual solar rebate!

For other articles in this series, see:
http://blog.dustinkirkland.com/search/label/Solar

:-Dustin

Going Solar in Austin, Texas


We're working hard to make your Ubuntu Server as energy efficient as possible, and I'm quite proud of that.

Personally taking it one step further, last week I signed a contract on a 6650W Photovoltaic System, to be installed on my roof while I'm in Dublin for the Karmic Distro Sprint.

The system consists of a total of 38 Solarworld 175 Watt Solar Panels, and a PVI-6000-OUTD-US Aurora Photovoltaic Inverter. It should supply the majority of our electricity (according to PV Watts) and pay itself off within a couple years (after rebate).

We contracted the system through my buddy Vincent Guerrero of Texas Solar Power Company, here in Austin, Texas, taking advantage of one of the most generous and progressive solar rebate programs in the country, offered by Austin Energy.

I figured I should blog our experience in case anyone else out there is considering the same. I'll keep you posted on the installation and performance of the system.

For other articles in this series, see:
http://blog.dustinkirkland.com/search/label/Solar

Cheers,
:-Dustin

Solar Installation - Part 1




So Step 1 in our solar project was upgrading the electrical panel. All of my breakers were spoken for, so I had to add a line side distribution box for the solar to interconnect. The junction box at the bottom allows me to add a sub-panel into my main distribution panel. The PV system will connect into this dedicated sub-panel.

Thanks to Joseph of Jackpot Electric for the nice, clean installation!

For other articles in this series, see:
http://blog.dustinkirkland.com/search/label/Solar

:-Dustin

Solar Installation - Part 2


Step 2 is installing aluminum brackets on the roof, to which the solar panels will actually mount.

I was chatting with one of the two installers. He noticed that I was wearing a Linux Foundation t-shirt, and he mentioned that he ran Linux on his older computer. Ubuntu, as it turned out. He said that he found it mostly user-friendly, and was able to do almost everything he could under Windows, except use Real Player to listen to his favorite college radio station, kdvs.org.

So I checked out kdvs.org, and found multiple m3u streams, for both mp3 and ogg formats. Within minutes, we were streaming to the outdoor speakers while they were working on the roof ;-) When he got home, he confirmed that he was able to do the same with Amarok.

For other articles in this series, see:
http://blog.dustinkirkland.com/search/label/Solar

:-Dustin

Solar Installation - Part 3

Alright, so with the rails up, we're ready for Part 3...installing the panels. The panels are mounted on the aluminum frames, and connected to a DC voltage collector, which feeds into the AC-DC power inverter.

We have a total of 38 modules going on the roof, in 3 separate "strings" or "arrays" (solar guy's terms). To those of us that are programmers:


power_t *solar1, *solar2, *solar3;
// versus
power_t solar1[16];
power_t solar2[11];
power_t solar3[11];


Heh, okay, bad joke!

A little over half are installed now. We're waiting on the shipment of the rest. Hopefully, the rest should be installed by the end of the week. Stay tuned...

For other articles in this series, see:
http://blog.dustinkirkland.com/search/label/Solar

:-Dustin


Solar Installation - Part 4





As of September 1, 2009, all 38 panels are now on the roof, and hooked up to the inverter, thought it's not yet turned on.

I had my inspection with Craig of Texas Solar Power Company, and David of Austin Energy today.

We spent about an hour going over the entire system. They turned it on for a few minutes, verifying the functionality. David went over every aspect of the design and installation. He noted a couple of things he wants to see updated in the CAD drawings and documentation. Craig is supposed to send those over to David as soon as possible, at which point my installation should be approved.

Now, I'm waiting 2 weeks for Austin Energy to come back out to my house to replace my current electric meter with one that can handle the PV junction. At that point, Austin Energy will actually turn on my system, and I will be generating Solar Power!

And then my rebate will be submitted for processing. I hope to see that rebate check in 6-8 weeks.

Many people have asked for more details about how the system works, and how my electric bill will be calculated.

The PV system will only generate electricity when the sun is out. Many factors will affect how much electricity is generated, including cloud cover, cleanliness of the panels, ambient temperature, length of the day, and the time of the year.

My system consists of 38 panels that are rated at 175 Watts apiece, for a theoretical maximum of 6650 Watts of output. The 38 panels are divided into two separate arrays. The Southwest-facing array consists of 3 strings of 9-panels apiece. These strings generate 357 Volts, 356 Volts, and 355 Volts apiece. Note the subtle difference in voltages. This ensures the direction of the flow (electricity always flows from higher voltage to lower voltage).

The amperage varies with the quality of the sunshine, and affects the output wattage. More sun yields more Amps, and thus more Watts. The other, Southeast-facing array consists of a single 11-panel string, which generates 439 Volts.

It is the job of the inverter to combine these two arrays and produce an AC current. My particular inverter is rated at 96.5% efficiency, so during the best hours of the day, I should see somewhere around 6400 Watts of electricity. During yesterday's test around Noon, with partial cloud cover, the system was generating 5550 Watts -- and this was far more electricity than I was consuming at that time.

The inverter is tuned to produce an AC voltage that is slightly higher than the power coming from Austin Energy's utility pole. This ensures that I use my solar generated power first, and any remaining flows back into the city's grid.

Now 6400 Watts is far more energy than I will be consuming during the daytime. However, at night, I won't be collecting any sunlight. During the day, the surplus will be used by others on the energy grid, and Austin Energy only buy that electricity from me at 40% of the price that I pay them for the electricity I use at night ;-) Such are the breaks... They pay the rebate, and thus they get to make the rules.

So we'll need to adapt our electrical usage model in order to maximize our savings. It will be in our best interest to run our laundry, dishwasher, and hot water heater during the daytime.

On a separate note, I'm quite eager to play with the serial data output on the Aurora inverter. Of course, my inverter came with a CDROM of Windows software that is worthless to me. Thankfully, someone out there has written a GPL application that communicates with the inverter's interface. I can't wait to play with it, once my system is turned on. Assuming it work's, I'll get it packaged and included in Ubuntu Karmic+1!

For other articles in this series, see:
http://blog.dustinkirkland.com/search/label/Solar

:-Dustin

October 29, 2009

Birds of Prey

I was on a conference call this morning when I spotted these two gorgeous birds of prey in the canyon below my house.

In this picture, you might see two tiny dots right in the center of the frame. This is about 500m away, with no zoom.


In the next frame, you can see the birds a bit more clearly, at 12x zoom.


With digital enhancement to 36x, you can see these are truly huge, beautiful birds.


Can anyone identify what type of bird they are? This is in Austin, Texas, for what it's worth. Perhaps a little too white to be an eagle or an osprey. A hawk or a falcon of some kind?

I think this was solved in the comments below... A red-tailed hawk, it must be!

:-Dustin

October 27, 2009

Palm Pre (WebOS) vs. HTC G1 (Android)




I blogged a couple of weeks ago about my new Palm Pre. I've been using it for over a month now, and I have a few more thoughts to share. But also, I won an HTC G1 from Qualcomm at the Linux Plumbers Conference in Portland. This was fortuitous, as I was quite interested in this device too.

I thought I would provide a comparison of the two devices, based on my experiences...

Network

Originally, I bought the Palm Pre because the G1 was not available on the Sprint network. While Sprint has disappointed and frustrated me more than once in the past 10 years that I have been a customer, their coverage in the USA is pretty good, and the unlimited data plans have served me well. I have not been under a contract in over 9 years, and don't intend on getting into a contract again with Sprint or any other carrier. I can use the Palm Pre in most of the USA, Canada, Mexico, and South America. But on the down side, I cannot use it in Europe or Asia.

The G1, on the other hand, takes a SIM card, and can be used on most any GSM network. I won a developer version of the G1, so it was unlocked by default. I simply borrowed a SIM card from a friend for 2 minutes, long enough to access the data network to login to my Google account and register the phone. After doing this, I gave Ted his SIM back, and my phone was functional as a stand alone "computer".

As it turns out, I do actually have a pre-paid SIM card from eKit.com, which I use in an ancient Siemens C60 phone when I'm traveling in Europe. While I have to use a second cell phone, I do prefer the prepaid route, as it's just too easy to spend several hundred dollars on your native carrier when traveling. So I was able to pop my eKit SIM card into my G1, and place and receive calls. I don't have a data plan on that SIM yet, so I wasn't able to test that.

HINT: To place a call using the eKit SIM on the G1, I needed to prepend *126* on the number I want to dial, and append #.

Connectivity

Both phones have WiFi and Bluetooth wireless connectivity, which is quite nice for large downloads. This, of course, absolutely drains the battery.

I mostly use the G1 as a WiFi device right now, since it's not connected to a cell service. It's great for browsing the web or using the various applications available on the platform (more about that below).

Most importantly, though, I have been able to make and receive calls over WiFi. There are many other blog posts floating around explaining how to do this. But basically I needed to:
  1. Apply for and receive a Google Voice account
  2. Register and activate a Gizmo5 account
  3. Add my Gizmo5 number to my Google Voice account
  4. Download and install SIP Droid on my G1
  5. Enter Gizmo5 username/password, and proxy01.sipphone.com as the server
  6. Enable WiFi, and wait for SIP Droid to authenticate (this part is buggy, as far as I can tell)
I have been able to do this well, from my home WiFi network. I'm looking forward to trying this abroad.

Tethering

The Palm Pre is a dream come true for anyone interested in tethering (though Sprint may apply additional charges to your account). With great pain, I have managed to tether using my last 3 Palm Treo's. With the Palm Pre and Ubuntu, however, it's ridiculously easy. I simply needed to root the device, and install My Tether. When I enable USB tethering, Ubuntu's Network Manager detects a new usb0 device, and grabs an address over DHCP. My practical download rate is an extremely respectable 150KB/s. I was able to maintain connectivity while riding passenger in a car over 7 hours and 450 miles between Austin, Texas and Baton Rouge, Louisiana last Thursday.

I have not been able to test the same operation on the G1 yet.

Console Access

I love that I can get a Linux shell on both of these devices. They both have small terminal applications accessible through the phone interface, but, of course, these keyboards are tiny and clumsy to do anything serious. Using vi, for instance, is a nightmare. But it's trivial to access either device through a terminal application from my Ubuntu desktop over the USB connection, and use a real keyboard.

I was able to install an SSH server on the Palm Pre, which is really nice, because I can now SSH directly to it, when it's connected on my WiFi network. No need to deal with cables or special terminal applications on my desktop.

Also, I like that I have root on the Palm Pre, and that it's possible to install packages from the command line. I have not yet obtained root on the G1. It looks slightly more difficult. I'll give that a shot soon.

Physical Device

The Pre is significantly smaller than the G1. It's easily the smallest cell phone I have ever owned, which is truly remarkable when you consider how powerful the device actually is.

I'm not a big fan of the little curve on the G1 toward the bottom. I think a completely flat device would be a bit more sleek.

However, the G1 feels like a very well made device. I like that you rotate the device sideways, when using the keyboard. The QWERTY keyboard is huge compared to any of the Palm or Blackberry devices. It's actually quite usable. Big keys, decent spacing. The hinge mechanism feels very sturdy, solid. I like the way it snaps into place, with no wiggle or play. I also really like the roller ball. There's some times I just don't want to touch my screen. Or I don't want someone else touching my screen. The roller ball, with the ability to go up/down/left/right and select is an excellent design feature.

Unfortunately, I think the physical design of the Palm Pre is its weakest point. The hinge feels very flimsy. It wiggles and has way too much play. It feels like it's about to break pretty much any time you're opening or closing it. It slides vertically, so, the screen is in portrait mode if you're using the keyboard. Most web pages and applications, it seems, perform better in landscape mode. The keyboard is also very small. The buttons don't rise as much as on the older Treos, and it takes a little while to get used to. I also very much hate that there is no up/down/left/right/select button. Everything has to be done with the keyboard and on screen gestures. This can be very painful for some applications, in my opinion. I'm really hoping the Pre is a "preview" of better hardware to come...

Ports

I love the fact that the G1's charging/syncing port is a very standard mini USB connection. I probably have 10 of these cables floating around, and it's fairly likely that if I don't, someone near me does, that I can borrow for a few minutes of juice.

The Palm Pre uses a strange little micro USB connector. Of course, I didn't have any of these cables, and one cable isn't going to cut it. So yes, I had to buy an extra cable or two (at $25 each). Disappointing. But, still, it's USB, and can charge nearly anywhere these days.

On the flip side, the Palm Pre has a standard 1/8" stereo jack. I can use my Palm Pre to play my MP3s or stream Pandora or internet radio stations directly to my stereo receiver, or to the auxiliary input in the car. I actually streamed 4 hours of the Saints/Dolphins football game on Sunday, driving back to Austin from New Orleans, pulling internet radio over the cell network and output to the car stereo. That was really cool ;-) And the Saints put together an amazing come-from-behind victory!

But the G1 uses the USB port to output sound too, which is odd. It came with a pair of stereo headphones, but I'll presumably need some sort of an adapter to send this signal to the car stereo.

It's disappointing that neither device quite gets the connection ports right. I want the Palm Pre's 1/8" audio jack, and the G1's standard micro USB jack in the same device. Is that too much to ask?

Storage

The Palm Pre ships with 8GB of storage built-in, but no expansion slot. The G1, on the other hand, has a microSD expansion slot. It was able to read my 16GB microSD card.

Seriously, Palm... This is really disappointing. The card and card reader are tiny. This omission is so obvious I conclude that this design decision is to ensure that Pre owners must upgrade their hardware to obtain more storage in the near future. I'm afraid this might well backfire on Palm, though.

My music collection has been under 80GB for almost 5 years now. At the rate that microSD cards are growing, I expect to buy a 128GB microSD card in the near future, and would hope to be able to carry my entire music collection on my mobile device. Unfortunately, it doesn't appear that the Pre will be that device...

Operating System

I'll probably devote an entire post to WebOS vs. Android, so I'll just hit the highlights here...

Both are Linux, Linux, Linux. That's just pure awesome. I've said it before... We have been hearing about Linux cell phones in the USA for years now, and that day is finally here.

Terminals on both--that's great. Both OSes are very stable and responsive. They both take a long time to boot, but the uptime has been excellent. I rarely need to reboot either of them.

The icons and effects on WebOS are perhaps a little sexier, but that's among the least of my concerns. I do find WebOS far more intuitive than Android. Finding applications, menus, settings, etc., is easier on WebOS.

Multi-tasking on WebOS is simply amazing. Launch program after program, moving it to the side, and bringing it back when you need it. It's ALT-TAB for your phone. I love it.

I see a lot of promise in both operating systems.

Upgrades

I have managed to upgrade the OS on both devices. The Palm Pre was very simple. Much like Ubuntu, I just checked for updates available, and applied those. I did so over the WiFi which was a bit quicker than over the cell network. Unlike Ubuntu, I couldn't easily see what was being updated. I'd like to find the equivalent of the ubuntu-changes mailing lists for WebOS...

Since I have the developer version of the G1, I was able to upgrade my OS to Android 1.5 (and I see that 1.6 is now available too). The upgrade procedure was straightforward, though definitely intended for developers, as you had to use the SDK to upload a new binary to the phone, and reboot into a special mode with an odd key combination. That said, this is fun for me, so I'm enjoying the G1 from the developer angle.

Applications

Android has been around a bit longer, and it seems, at least at this point, to have a larger developer community. The Android "Market" has more applications than the WebOS "App Catalog", just from a numbers perspective. But there are a lot of duplicates. I have found almost all of the key applications I need for my Pre (with the notable exception of a SIP client, though it appears that one is under development).

It's clear that Apple's iPhone has a tremendous advantage on the App Store front. I think the rest of the cell phone OS market would do well to converge a bit, and confront Apple together.

That said, I would really, really love to have access to Ubuntu's archive of 20,000+ applications on my Linux devices. It seems like such a rich resource to tap into. Most of these devices have ARM processors, and we're actively working on Ubuntu's ARM story. With the 8GB of space on my Palm Pre, I could easily apt-get install most of the applications I carry around on my desktop. I'd like to think we're not too, too far away from that day...

Accessories

I don't collect a lot of accessories. Actually, I don't have any accessories for the G1.

I did buy 2 Palm Touchstones, on the premise that this technology was pretty cool. It's a little hockey puck sized "wireless charger". It uses magnetic induction to charge the Pre. When it works, anyway. I've run into a number of problems, most of which seem to be software. It seems that the software side doesn't always detect that the phone is sitting on the Touchstone. More importantly, it seems that if the software doesn't detect that it's on the charger, it doesn't start charging. I have run my Pre completely out of battery more than once now, and when this happens, the Touchstone won't do you any good. You have to plug into the USB. This creates a vicious cycle, if you put the Pre on the Touchstone, but it doesn't charge, it runs out of battery, and then you have to plug into the USB to get it usable again. I must say that I'm extremely disappointed with the Touchstone. I think I'm going to try and return them. This issues might be fixed in a software update, I suppose, but right now, the Touchstone is not worth the already-overpriced $70 tag.

Conclusion

I'm definitely enjoying both devices. Linux on the phone has a promising future. WiFi, QWERTY keyboards, App Stores, these are all great things.

The Palm Pre would be the perfect device if it had:
  • roller ball for up/down/left/right/select
  • removable microSD card slot
  • sturdier sliding mechanism
  • standard mini USB connector
  • decent SIP client for VOIP calls
  • longer battery life
The G1 would be the perfect device if it:
  • were slightly smaller, lighter
  • were available on the Sprint network
  • had a stereo audio output
  • had longer battery life

:-Dustin

Server Team 20091026 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

  • kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt: in progress

ACTION: kirkland to finish adapting help.ubuntu.com VM recipe(s?) to use libvirt

  • mathiaz to add test case for image store in testcases wiki: Done

  • zul to write up server upgrade test case: Done

  • nurmi to investigate bug 455625: TODO

ACTION: nurmi to investigate bug 455625

  • soren to complete demo virtual appliance: Published and tested, could use some polish
  • mdz to chase down details on production image store: Running and tested
  • mathiaz to test UEC integration with production image store: Tested and working OK

Karmic release last actions

Remaining RC bugs:

  • 458850: on track to be included on release cloud images

  • 458576: is safe to fix and would get rolled in the same respin

ISO testing and mentioning relevant bugs in the release notes should be the focus of this week.

Release notes candidates are:

  • 459101: Relay denied from eucalyptus registration emails

  • 444352: DB deadlock on reboot prevents UEC from working, temporarily

kirkland will review the bug lists and mention anything else UEC-related that I think is worth release-noting.

We should also help where we can the Foundations team in solving 457767 (iSCSI boot). It now seems related to kernel issue so wee should update the documentation to reflect current support.

Lucid Lynx

mdz gathered requirements from Canonical stakeholders, with some more to come. Items are expected in the same broad categories as 9.10: UEC, EC2, appliances, enterprise deployment items. But we will aim to focus on stabilization and negotiate our commitments with regard to that. Opening the meeting to 10.04 suggestions, lots of ideas were discussed. A wikipage was created to followup on those suggestions (and allow to suggest more): https://wiki.ubuntu.com/ServerTeam/LucidIdeaPool

ACTION: ttx to create wiki page for 10.04 input from server team (Done)

Open discussion

kirkland mentioned that the new time slot for the meeting (Wednesdays at 1400 UTC) is difficult for him, so we might (again) change it. ttx asked about how the triage days were going, and updated the knowledge base to make information about them more accessible. mdz asked about release parties that the team plans to go to, as a great way to celebrate and let the pressure go. mathiaz plans to get free beers all night.

Agree on next meeting date and time

Next meeting will be on Wednesday, November 4th at 14:00 UTC in #ubuntu-meeting.

October 26, 2009

Linux Magazine: Ubuntu Encrypted Home

Back in April, Linux Magazine ran what I considered to be an inaccurate account of the OS-level security provided by our Ubuntu Distribution. Your Distro is Insecure: Ubuntu.

Frustrated with the piece, I blogged this in return: Your Article is Incorrect: Linux Magazine.

Following that post, I had a very constructive, private email conversation with Linux Magazine editor, Bryan Richard. We discussed a number of different ways that Canonical/Ubuntu might be able to respond to their previous article, which caused quite a stir on Ubuntu's public mailing lists.

I'm very pleased with Bryan's response. He invited me to author an article focusing on the security features that are available in Ubuntu. The result was published earlier today, focusing on Ubuntu's Encrypted Home Directory feature, which is rather unique among Linux distributions: Ubuntu's Encrypted Home Directory: A Canonical Approach to Data Privacy


Enjoy!
:-Dustin

Austin's Karmic Release Party!


Thanks to everyone who joined us at Austin's Jaunty Release Party in April 2009, which we held at 6th's Jackalope Bar -- oh so perfectly named!

I had hoped to follow up our synchronicity with the Karmic Release Party at a bar called Karma in downtown Austin. Unfortunately, it doesn't open until 9pm, which is probably a bit late for our crowd. So I needed to find another interesting, appropriate venue.

Join us between
6pm - 9pm on Thursday, October 29, 2009 at Aussie's Grill and Beach Bar, at 306 Barton Springs Road, just east of the corner of Riverside and Barton Springs Road, south of the river. I figured an Australian-themed bar would have to work, in honor of the home of our lovely mascot, the Karmic Koala!

Burning CD's is so '90s... If you'd like a bootable/installable copy of Ubuntu 9.10 Karmic Koala, bring an empty USB stick, 1GB or bigger, and we'll gladly burn an image for you. There should be a couple of laptops demoing the new release. I'm planning on bringing two, to demo the Ubuntu Enterprise Cloud (UEC), powered by Eucalyptus, which I've been working on for the last few months.

Also, leave a note in the comments below, if you plan on joining us, so I'll have some idea of how much space to reserve. Look for us outdoors, on the patio (weather permitting). And perhaps we'll do a little after party at Karma, just to say we did! :-)

:-Dustin

October 25, 2009

TPM as RNG

I was reminded about some TPM coding I’d done to get random bytes from the pRNG on my TPM-enabled system from Matt Domsch’s recent post. I’m not fully convinced that the pRNG of the TPM is an appropriate source of entropy, but it does pass my simple FIPS-140-2 test.

I had to find the Intel TPM docs to figure out how to enable TPM on my system. It was under “Advanced / Peripherals”. I was expecting it under “Security”, like every other BIOS I’d seen. After that:

$ sudo apt-get install trousers tpm-tools
...
$ sudo modprobe tpm_tis
$ dmesg | grep -i tpm
[676618.167313] tpm_tis 00:07: 1.2 TPM (device-id 0xFE, rev-id 70)
$ sudo service trousers start
...
$ tpm_version
TPM 1.2 Version Info:
Chip Version: 1.2.2.16
Spec Level: 2
Errata Revision: 1
TPM Vendor ID: WEC
TPM Version: 01010000
Manufacturer Info: 57454300
$ ./tpm-getrand | hexdump -C
00000000 61 07 23 ff 71 3e 25 e8 f0 d5 de a7 a3 07 21 dc |a.#.q>%.......!.|

I could run rngd with a named pipe, but it’d be nice to have a new driver that could run a command instead to get the next 20000 bits.

UPDATE: I’ve implemented this in rngd now.

October 24, 2009

karmic and log rotation

In Ubuntu’s Karmic and and Debian’s Lenny, sysklogd was replaced with rsyslog. This is fine, since rsyslog will have converted your /etc/syslog.conf to /etc/rsyslog.d/50-default.conf. However, if you modified the (maddeningly strange sysklogd-specific) log file rotation in /etc/cron.daily/sysklogd or /etc/cron.weekly/sysklogd, you’ll want to review the new (sane) /etc/logrotate.d/rsyslog. (Note also that savelog uses .0 as the first rotated file extension, and logrotate uses .1.)

October 21, 2009

Ubuntu Virtualization Poll - Your Feedback Requested!


We're still a week away from releasing Ubuntu 9.10, which I'm sure will be a phenomenal server release, with huge strides in virtualization and cloud hosting. The Ubuntu Enterprise Cloud should be the most complete open source cloud hosting solution in the industry.

But we're also beginning to prepare for the Ubuntu Developer Summit in Dallas, Texas next month. At this summit, we will discuss our plans for Ubuntu Lucid Lynx, which will release in April 2010 as Ubuntu 10.04 LTS. This being an LTS release, UDS is incredibly important, as these decisions will affect the Ubuntu landscape for at least 5 years.

As your maintainer of Ubuntu's virtualization stack supported by Canonical, I'm pleased to invite you to provide feedback on virtualization in Ubuntu in this simple, brief, 6-question survey:


We are eager to hear your feedback on a few particular questions about KVM, QEMU, Virsh, Virt-Manager, Xen, VirtualBox, OpenVZ, VMWare, Parallels, Amazon EC2, Eucalyptus, and other virtualization technologies.

Note: Nick Barcet will be conducting a much more comprehensive Ubuntu Server Survey in the near future. Stay tuned!


Thanks!
:-Dustin

Server Team 20091020 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

  • kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt: Deferred

ACTION: kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt

  • zul to fix m2crypto test suite and ensure that MIR is processed: Done
  • mathiaz to add test case for image store in testcases wiki: Not done yet

ACTION: mathiaz to add test case for image store in testcases wiki

  • zul to add missing server-ship packages to ubuntu-server: Done

Karmic RC

Review of remaining karmic-nominated bugs:

  • 455832: Not a regression and not encountered in normal use: untargeted for release

  • 451881: Only affects UEC, and not EC2, not critical for karmic: untargeted for release

  • 410886: vmbuilder needs to have the patch from trunk merged into it: milestoned and assigned to soren

  • 362511: Low importance, won’t get fixed for karmic: untargeted for release

  • 453456 and 455293: see below

  • 455411: Harmless to users, too late to fix: untargeted for release

  • 453495: Specific to qemu-only arches, can be dealt with postrelease: untargeted for release

  • 453453: doesn’t seem realistic to do anything about this bug for 9.10: left at kirkland’s discretion

  • 453467: not 9.10-critical, untargeted for release

  • 407949: milestoned to release, assigned to smoser, will trigger a UEC/EC2 image respin

Review of the Ubuntu-Server-relevant tests on the ISO tracker revealed a few test cases that were not covered by any team member:

  • netboot: Daviey and kirkland will cover those, as they have a local mirror
  • upgrades: zul will cover those cases. This involves defining a test procedure, as the current one is Desktop-oriented.

ACTION: zul to write up server upgrade test case

Eucalyptus status

We fixed ~36 bugs in 9 uploads last week. A few remaining bugs were mentioned during the review:

  • 455816: Given that this is globally harmless and difficult to test between RC and release, it’s unlikely that this will be fixed for release. mathiaz will validate the fix though.

  • 453456: Would be fixed only if log rotation is proved to be non-functional, and can be fixed in a SRU

  • 455293: Fix is committed but is not worth a new upload. Will get in only if something else gets fixed

  • 455625: This bug seems bad, though it still needs to be reproduced. Would get fixed in a Karmic SRU. nurmi is looking into reproducing it.

  • 452556 and 444352: ttx hits those regularly in ISO testing, the first one would warrant a note in release notes, this it seems systematic (task opened)

ACTION: nurmi to investigate bug 455625

UEC/EC2 images

This was reviewed separately and found to be in good shape.

Reference UEC appliance

The demo will very closely resemble what soren posted the week before last. The completely unpersistent appliance will be ready by Thursday.

ACTION: soren to complete demo virtual appliance

UEC appliance store

Integration with the “fake” test store was successfully tested. This needs now to be tested ASAP against the production server, if available. It can be tseted against Soren’s demo appliance but also the stock UEC image.

ACTION: mdz to chase down details on production image store

ACTION: mathiaz to test UEC integration with production image store

UEC documentation

Current community doc (http://help.ubuntu.com/community/UEC and http://help.ubuntu.com/community/UEC/PackageInstall) needs to be worked on. kirkland, nurmi, mathiaz and ttx stepped up to help in that area. Discussion on splitting the work amongst this team will be held on a mailing-list of kirkland’s choice.

Assigned and to-be-assigned bugs review

455873 was assigned to zul, but since it is for hardy, release-related tasks should take precedence.

AOB

ScottK mentioned that clamav 0.95.3 will likely be released on Monday, so it’ll go to -proposed.

Agree on next meeting date and time

Next meeting will be on Tuesday, October 27th at 15:00 UTC in #ubuntu-meeting.

October 15, 2009

Really Enjoying Pandora

So I tried Pandora a few years ago, but initially wasn't that impressed. And besides, I had developed Musica, a web application for browsing and streaming my own music over the web.

In May, I received a Chumby as a gift from Canonical. Five months later, I use it almost exclusively as a Pandora front end, attached to a nice set of powered Klipsch speakers.

It took a few minutes, but I entered about 40 of my favorite artists, and now use a QuickMix of those artists plus other similar artists.

They've done a great job creating applications for lots of platforms. I can stream Pandora now through my:
  • Ubuntu computers
  • Chumby
  • Palm Pre
  • G1
  • Samsung Blu-ray
I finally decided to shell out the $36 for PandoraOne, and I've been very satisfied. The extra bitrate, lack of commercials, and unlimited play time have been very nice.

I still use Musica when I want to listen to something very specific. But Pandora has been really nice for an eclectic mix.

On the down side, it appears that Pandora is only available to IP addresses in the USA. Apologies to readers outside of the USA :-(

:-Dustin

October 14, 2009

Server Team 20091013 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

  • ttx to review test plans and ensure they are aligned with 9.10: Done, pending some alignment by ara and the QA team
  • kirkland to confirm that his test rig is fully operational: Done
  • kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt: Deferred

ACTION: kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt

  • mathiaz to work with the QA team on a server bug day for Karmic: Done, running Oct 14
  • mdz to follow up with marjo regarding general QA support in Karmic: Done
  • smoser to follow up with mdz regarding UEC image testing capability: HW received, setup to be done
  • zul to fix m2crypto test suite and ensure that MIR is processed: some more fixups needed to testsuite

ACTION: zul to fix m2crypto test suite and ensure that MIR is processed

  • soren/niemeyer to arrange a meeting to discuss reference appliance plan of action: Discussed via email
  • mathiaz to document test plan for image store: niemeyer wrote instructions in image-store-proxy README

ACTION: mathiaz to add test case for image store in testcases wiki

  • zul to triage his assigned bugs: Done

Eucalyptus status

Eucalyptus is currently at 1.6~bzr919-0ubuntu3. We expect a last upstream merge today, and it’s likely to be > rev925 so we should wait for it before merging (fixes for 430266 and 449944). It should be delivered before 5pm PST on Tuesday. 432154 is still stuck on integration of an upstream patch that might fix SCSI attach. 446023 is also on kirkland’s tasklist and should get fixed asap. 446841 needs some upstream help which nurmi should provide today.

UEC/EC2 images

All serious bugs have now been squashed. 440757 is next on smoser’s list. 444605 should now be marked as completed, since slangasek agreed that with the -kernel-info.txt file and even just the manifest we have what we need.

Server Team bugmail

zul went through the server-ship seeds to make sure that the server team is subscribed to the bugmail for the stuff in the seeds and came up with a list of missing items. After some minor cleanups (packages no longer in the archive that should be removed from the seed), we agreed that this list should be added to ubuntu-server package list.

ACTION: zul to add missing server-ship packages to ubuntu-server

RC – FinalFreeze ahead

FinalFreeze is coming up this Thursday, so most fixes should be in by Wednesday. Only release-critical fixes will be granted an exception after that date. A server bug day will be held Wednesday to help catching the missing bugs that should be targeted to release. See https://wiki.ubuntu.com/UbuntuBugDay/20091014 for more information.

Agree on next meeting date and time

We are still trying to find the best time for this meeting. Something along the lines of what the desktop team uses (https://wiki.ubuntu.com/DesktopTeam/MeetingTime) could be used to help in the process.

Next meeting will be on Tuesday, October 20th at 15:00 UTC in #ubuntu-meeting.

October 13, 2009

It is quiet in here?

Dear Blog,

I am really sorry, I've been unfaithful. I have been seing another blog lately. Please don't be mad...

I have accepted to write a monthly blog on WorksWithU.  A couple entries already made:

read more

libcgroup in Ubuntu Karmic

Earlier in the Karmic cycle, I worked with Jon Bernard to sponsor his packaging of libcgroup, an open source library from IBM that provides tools for controlling and monitoring control groups.

In this post, Jon teaches you how to use libcgroup to your advantage. He has a work station that is his primary desktop, but also running Apache2, as well as a valid system user named kirkland that's chewing up his CPU cycles. Jon's solution is to put kirkland and all processes created by that yahoo into his own cgroup, and only receive CPU cycles when no other processes are in need. While this is a simple, silly demonstration, Ubuntu Server administrators should find this feature useful in Ubuntu 9.10!

Enjoy!

http://jonbernard.blogspot.com/2009/10/ubuntu-resource-management-very-simple.html

:-Dustin

Run your own Ubuntu Enterprise Cloud, part 3


In part 1 and part 2 of this series, we saw how to set up a minimal cloud infrastructure and bundle a basic image (and test it). In this final article, we’ll play with our cloud from an end-user perspective.

Setting up the web UI

First of all, before accepting end users, as the administrator of the cloud you will have to setup a few things on the web UI. Using your favorite browser, you should:

  • Open https://cloudcontroller:8443/
  • Log in using the default user/password: admin/admin
  • Change the default password, setup the cloud admin email address
  • Logout

Setting up the cloud client

We’ll use Ubuntu 9.10 beta for this purpose, as it includes all the needed packages, and it’s so great ! You will have to install the following packages:

$ sudo apt-get install euca2ools unzip
 

Registering on UEC, getting credentials

As the end-user, fire up your favorite browser and:

  • Open https://cloudcontroller:8443/
  • Click “Apply” and enter your end user details

If you set up the email correctly on your cloud controller, it should send an email to the cloud admin address asking him to approve that request. Follow the instructions on that email to approve the account as the admin.

You should then get an email at the end user email address asking you to confirm the account request. Follow the instructions on that email, then you can log in on the web UI:

  • Open https://cloudcontroller:8443/
  • Login using your end user username and password
  • Click “Download Credentials” in the “Credentials” tab
  • Note the EMI reference you can use on the “Images” tab

Starting up an instance

You should unzip the credentials zipfile you just downloaded, then source the eucarc file and test the connection:

$ unzip euca2-enduser-x509.zip
$ . eucarc
$ euca-describe-availability-zones verbose
 

Setup a SSH key and allow connection to the SSH port:

$ euca-add-keypair enduserkey > enduserkey.priv
$ chmod 0600 enduserkey.priv
$ euca-authorize default -P tcp -p 22 -s 0.0.0.0/0
 

Then starting up an instance is just a matter of passing the right EMI and type:

$ euca-run-instances -k enduserkey emi-XXXXXXXX -t c1.medium
 

Enjoy !

October 08, 2009

Server Team 20091006 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

  • niemeyer to send mathiaz and nurmi a mail with details on how to test the image-store-proxy integration with fakestoreapi.py: Done, though not tested yet
  • ttx and smoser to review the automation of the image publication process: Done. Some automation steps are still in progress, should be completed by RC
  • smoser to open bugs to cover kernel/ramdisk GPL reqs and renaming: Done, bugs 444598 and 444605
  • ttx to test UEC images + UEC kernel/ramdisk on karmic UEC: Done for beta release
  • ttx to see if some bugs assigned to soren need urgent reassignment: Done but needs some update now
  • Daviey to setup a doodle poll and send the url to the ubuntu-server@ mailing list: Done, but suboptimal

Beta release postmortem

ttx reported that beta release went relatively ok, the main issue was lack of UEC testing capability, which made it difficult to confirm the bugs that were found. mdz pointed at a blind spot in testing with regard to the UEC kernel and initramfs, which might require some adjustments on the test plan. Further UEC feature testing should now use the UEC beta image, together with published kernel and ramdisk. The need for at least two complete UEC testing setups in different timezones was mentioned. mathiaz should investigate the possibility of using some hardware from the certification labs. A reference Canonical UEC setup running on 1.6 should be the right place to validate UEC images.

ACTION: ttx to review test plans and ensure they are aligned with 9.10

ACTION: kirkland to confirm that his test rig is fully operational

RC preparation

Status on a few actions:

  • Automate publishing of AMIs to EC2 [smoser]: in progress
  • Publish ec2-version-query in a more appropriate place, automation [soren]: now in smoser hands
  • mathiaz to set up OpenLDAP/sssd test infra on EC2: in progress
  • mathiaz to get help from Michael Vogt on bug 194140: mvo will look at it tomorrow
  • kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt: still todo, planned an OpenWeek session about it

ACTION: kirkland to adapt help.ubuntu.com VM recipe(s?) to use libvirt

With FinalFreeze coming up on October 15th, we need to quickly identify missing release-critical issues and regressions in server packages. mdz stressed again the importance of using the regression-potential tag in that process, see https://wiki.ubuntu.com/QATeam/RegressionTracking for more details. kirkland mentioned libvirt as one package requiring a complete triage effort, quickly organizing a bug day on that package could help. For other packages, zul and mathiaz will be coordinating the effort of sweeping through the bug lists: mathiaz will look at Confirmed/Triaged bugs, while zul will look after New/Undecided ones. Other team members should help in that effort, especially for the packages that they usually take care of. A server bug day should be organized to provide a framework for community participation.

ACTION: mathiaz to work with the QA team on a server bug day for Karmic

ACTION: mdz to follow up with marjo regarding general QA support in Karmic

Parallel to that effort is the need to stay on top of new bugs filed during Beta testing, to quickly identify the potential show-stoppers. A process has been defined to at least set the importance of NEW bugs, and a team member was assigned to each day (to take care of the bugs filed during the day before). The backlog should be taken care of progressively.

Eucalyptus status

kirkland uploaded a new eucalyptus yesterday, merge of an upstream bugfix revision that closes ~12 bugs. ttx tested it, autoregistration still needs another pass as it seems to work by accident in that version (bugs 443325,444504), and a deadlock bug still needs some investigation (444352). kirkland holds a UEC-hardening mini-sprint with mathiaz and nurmi in Austin this week, where bug 432154 and 436977 should be discussed.

UEC/EC2 images

Reviewing the UEC image bug list: smoser will target to release appropriately (bugs 444605 444598 440757), and will take some time to setup a UEC image testing environment. zul should make sure the last MIR gets completed, by fixing the testsuite.

ACTION: smoser to follow up with mdz regarding UEC image testing capability

ACTION: zul to fix m2crypto test suite and ensure that MIR is processed

Reviewing the EC2 images buglist: jjohansen has put up some kernels with tweaked configs (bug 428692). Note that it is possible to test UEC images in plain KVM by passing kernel parameter ec2init=0. ttx and kirkland will dedicate some of their UEC testing time testing the candidate UEC images on smoser’s request.

Virtual appliance

The final design questions around the reference appliance need to be resolved ASAP.

ACTION: soren/niemeyer to arrange a meeting to discuss reference appliance plan of action

On the imagestore proxy side, niemeyer will provide mathiaz with an updated version with a couple of very minor changes today. Testing instructions are in mathiaz’s hands, and need to be documented in a test plan, that the happy few with a UEC setup can help test.

ACTION: mathiaz to document test plan for image store

Assigned bugs

mdz pointed to importance-Undecided bugs on the list, mostly assigned to zul, that need to be cleared. The highest importance triaged bugs on that list are the vmbuilder ones, assigned to soren, which should dedicate some time to vmbuilder tomorrow.

ACTION: zul to triage his assigned bugs

Agree on next meeting date and time

Discussion over a better time for the team meeting should be held on the mailing-list. Everyone should answer there.

Next meeting will be on Tuesday, October 13th at 15:00 UTC in #ubuntu-meeting.

October 07, 2009

Run your own Ubuntu Enterprise Cloud, part 2


In part 1 of this series, we saw how to install the cloud infrastructure. In this article, we’ll bundle and upload an EMI (Eucalyptus Machine Image), based on Ubuntu Server 9.10 Beta, and validate that we can run an instance of it.

Download required elements

Go to the cloud/cluster controller and download the required items.

For a 64-bit image:

$ URL="http://uec-images.ubuntu.com/releases/karmic"
$ wget -O image.gz $URL/beta/ubuntu-uec-karmic-amd64.img.gz
$ wget -O vmlinuz $URL/beta/ubuntu-uec-karmic-amd64-vmlinuz-2.6.31-11-server
$ wget -O initrd $URL/beta/ubuntu-uec-karmic-amd64-initrd.img-2.6.31-11-server
 

For a 32-bit image:

$ URL="http://uec-images.ubuntu.com/releases/karmic"
$ wget -O image.gz $URL/beta/ubuntu-uec-karmic-i386.img.gz
$ wget -O vmlinuz $URL/beta/ubuntu-uec-karmic-i386-vmlinuz-2.6.31-11-generic-pae
$ wget -O initrd $URL/beta/ubuntu-uec-karmic-i386-initrd.img-2.6.31-11-generic-pae
 

Bundle the EMI

First you should unpack and resize your image to the desired size, lets say 4Gb. This can take a very long time (15 minutes !) on slow disks as you unpack 10Gb-worth of image space:

$ zcat -f image.gz | cp --sparse=always /dev/stdin image
$ e2fsck -f image
$ resize2fs image 4G
$ truncate --size=4G image
 

Then bundle and upload the kernel:

$ . eucarc
$ euca-bundle-image -i vmlinuz --kernel true
$ euca-upload-bundle -b ueckernel -m /tmp/vmlinuz.manifest.xml
$ euca-register ueckernel/vmlinuz.manifest.xml
IMAGE     eki-KKKKKKKK
 

Take note of the EKI reference, you’ll need it later. Then bundle, upload and register the ramdisk:

$ euca-bundle-image -i initrd --ramdisk true
$ euca-upload-bundle -b uecramdisk -m /tmp/initrd.manifest.xml
$ euca-register uecramdisk/initrd.manifest.xml
IMAGE     eri-RRRRRRRR
 

Take note of the ERI reference. Finally, bundle the image with the kernel and ramdisk, upload and register:

$ euca-bundle-image -i image --kernel eki-KKKKKKKK --ramdisk eri-RRRRRRRR
$ euca-upload-bundle -b uecimage -m /tmp/image.manifest.xml
$ euca-register uecimage/image.manifest.xml
IMAGE     emi-XXXXXXXX
 

Bundling will also take a lot of time ! Take note of your EMI reference.

Start an instance of your EMI

In order to access your instance using SSH, you’ll need to setup a few one-time things (create a SSH key and authorize access to port 22 of your instances):

$ euca-add-keypair mykey > mykey.priv
$ chmod 0600 mykey.priv
$ euca-authorize default -P tcp -p 22 -s 0.0.0.0/0
 

Now it’s time to start your instance !

$ euca-run-instances -k mykey emi-XXXXXXXX -t c1.medium
 

The “c1.medium” VM type is sufficient by default to run a 4Gb instance. You should take note of the i-YYYYYYYY reference that is displayed on your INSTANCE line. The first time you start an EMI, it can take some time (like 10 minutes) to move from “pending” state to “running”, depending on size. You can use the following command to automatically watch the output of euca-describe-instances, every 5 seconds:

$ watch -n 5 euca-describe-instances
 

Take note of the first ZZZ.ZZZ.ZZZ.ZZZ IP address mentioned in the output of the command. When the instance is “running”, ctrl-C to exit watch, then:

$ ssh -i mykey.priv ubuntu@ZZZ.ZZZ.ZZZ.ZZZ
 

You are in ! When you’re done playing with your instance, just run the following command on the cloud/cluster controller.

$ euca-terminate-instances i-YYYYYYYY
 

In the third and last part of this series of articles, we’ll talk about how to run instances from another workstation, as a cloud “customer”.

October 04, 2009

Run your own Ubuntu Enterprise Cloud, part 1


Ubuntu Enterprise Cloud is the product, powered by Eucalyptus, that allows you to easily run your own Amazon-EC2-like private cloud. It’s a lot simpler than you’d think. With the recent Ubuntu Server 9.10 beta release, you are now able to easily deploy that infrastructure from the CD installer.

Prerequisites

To deploy a minimal cloud infrastructure, you’ll need at least two dedicated systems. One will hold the cloud controller (clc), the cluster controller (cc), walrus (the S3-like storage service) and the storage controller (sc). This one needs fast disks and a reasonably fast processor. The other system(s) are node controllers (nc) that will actually run the instances. These ones need CPUs with VT extensions, lots of CPU cores, lots of RAM, and fast disks. For both, 64-bit support is highly recommended.

Installing the cloud/cluster controller

Download the 9.10 Server beta ISO. When you boot, select “Ubuntu Enterprise Cloud install”. When asked whether you want a “Cluster” or a “Node” install, select “Cluster”. It will ask two other cloud-specific questions during the course of the install:

  1. Name of your cluster: pick any name you want :)
  2. List of IP addresses on the LAN that the cloud can allocate to instances: enter a list of space-separated unused IP addresses on your LAN.

When it reboots, run the following to get the latest eucalyptus package and reboot:

$ sudo apt-get update
$ sudo apt-get upgrade
$ sudo reboot
 

Installing node controllers

The node controller install is even simpler. Just make sure that you are connected to the network on which the cloud/cluster controller is already running. Take the same ISO, select “Ubuntu Enterprise Cloud install”. It should detect the Cluster and preselect “Node” install for you. That’s all.

It is also recommended to update to the latest 9.10 status:

$ sudo apt-get update
$ sudo apt-get upgrade
 

Connect your node controllers to the cloud

After all nodes are installed, you need to return to the cloud/controller and run the following command to make it “discover” your newly-installed nodes.

$ sudo euca_conf --no-rsync --discover-nodes
 

Confirm all the nodes it finds, and you are done. To check that your private cloud infrastructure is ready to serve, you need to retrieve admin credentials and run euca-describe-availability-zones command. Run the following on your cloud/cluster controller:

$ sudo euca_conf --get-credentials mycreds.zip
$ unzip mycreds.zip
$ . eucarc
$ euca-describe-availability-zones verbose
 

This last command returns a description of the capabilities of your cloud cluster, how many instances of each type you could run on it, for example:

AVAILABILITYZONE   myowncloud                 192.168.1.1
AVAILABILITYZONE   |- vm types                free / max   cpu   ram  disk
AVAILABILITYZONE   |- m1.small                0004 / 0004   1    128     2
AVAILABILITYZONE   |- c1.medium               0004 / 0004   1    256     5
AVAILABILITYZONE   |- m1.large                0002 / 0002   2    512    10
AVAILABILITYZONE   |- m1.xlarge               0002 / 0002   2   1024    20
AVAILABILITYZONE   |- c1.xlarge               0001 / 0001   4   2048    20
 

In part 2 of this series, we’ll cover bundling your first EMI (Eucalyptus Machine Image), based on Ubuntu Server 9.10 Beta. We’ll test it by starting an instance of it. Stay tuned !

October 01, 2009

Server Team 20090929 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Virtual appliance

kirkland mentioned he moved away from a moodle to a gobby server appliance. He also added that we needed to figure out how to make appliances out of packages that had critical debconf questions. It’s a generic problem and requires admin input on package installation. For the appliance to be useful all configuration should be web driven after instantiation. He asked whether the debconf web frontend could be used for that. cjwatson acknowledged that there was one although it wasn’t very good.

niemeyer reported that the proxy was in a pretty good state now. Work on the server side is required and he is waiting on the server infrastructure for this. He also noted that the server side can be faked with fakestoreapi.py for testing purposes.

ACTION: niemeyer to send mathiaz and nurmi a mail with details on how to test the image-store-proxy integration with fakestoreapi.py.

Asterisk stack

Daviey reported that Asterisk 1.6rc2 had been uploaded to karmic.

vmbuilder

smoser reported that vmbuilder used the seeds to build the list of packages to include in the UEC images. mdz asked that vmbuilder’s version be included in the manifest. This turns out to be less straightforward as vmbuilder trunk branch is currently used to build images.

Automate image publishing and ec2-version-query refresh

smoser reported that the image publication process was somehow automated. He started to write some scripts. The goal is moving towards being ready to work right away on image publication automation as soon as we get some access to amazon from data center.

ACTION: ttx and smoser to review the automation of the image publication process.

Roadmap review / Eucalyptus

kirkland put a huge effort into triaging eucalyptus bugs – the list is usable again.

ttx mentioned that the Eucalyptus help page was outdated especially in the Installer integration section. He asked for help on refreshing the content based on the test cases. alexm offered to review the doc while Daviey will try to update parts of it.

ACTION: kirkland to fix bug 438747, maybe bug 438602, package and push a respin with that package in

Roadmap review / UEC – EC2 bugs

smoser reported that all bugs targeted to beta with ec2-images or uec-images in a tag were fixed as far as he knew.

zul mentioned that amazon had released a new version of ec2-api-tools which added better ebs and vpn support. ttz suggested it was probably too late for the karmic release cycle. zul could just make them available in a ppa and upload it for lucid.

ACTION: smoser to open bugs to cover kernel/ramdisk GPL reqs and renaming

ACTION: ttx to test UEC images + UEC kernel/ramdisk on karmic UEC

Assigned and to-be-assigned bugs

ACTION: ttx to see if some bugs assigned to soren need urgent reassignment

Weekly SRU review

mathiaz conducted the weekly SRU review and relevant bugs have been accepted. He also reminded that the updated SRU workflow in the Ubuntu Server team was using package branches to prepare SRU. sbeattie mentioned he was interested in participating at the review stage making sure that both the package changes (via a merge proposal) as well as the SRU report were in good shape before the upload to the archive.

2009 Server survey status and call for action

nijaba announced he had updated the 2009 Server survey. A test server is available for reviewing the updated questions. Bugs should be filed in LP against the server-survey project.

Agree on next meeting date and time

soren started a discussion on the ubuntu-server mailing list about a new time and date for the meeting. Daviey suggested to use doodle to conduct a poll.

ACTION: Daviey to setup a doodle poll and send the url to the ubuntu-server@ mailing list.

Next meeting will be on Tuesday, October 6th at 15:00 UTC in #ubuntu-meeting.

A Hacker's Drive Across Scotland


In August 2008, I spent a week in London for an Ubuntu Sprint, and then took a week of vacation to hike nearly 70 miles across Scotland, along the Speyside Way. Before I had even left Scotland, I was already planning my next trip!

To my good fortune, the Ubuntu Platform Sprint was held in Dublin in August 2009, and I found myself once again on the East side of the Atlantic. Now Dublin is a positively lovely city. Wonderful people, very walkable, safe, old, and beautiful. Guinness does in fact taste better in on the Emerald Isle. And the cool days and cooler nights are a welcome relief from the blazing August heat in Austin. We got quite a bit of work done, polishing up some of Ubuntu's Karmic Koala Server features (and saving some others for what will likely be a photo finish!). We toured the Guinness Factory (of course), and had a few pints in the Gravity Bar overlooking all of Dublin. The movie Once is easily one of my favorites of all time, and it was great to see so many of the settings from the film. I met up with my friends Josh and Erin for a tour of their neighborhood and visited their apartment and rooftop views. On the recommendation of my very Irish buddy, Eddie, I visited Gogarty's in Temple Bar, and sampled a few Irish whiskeys, including Green Spot and Powers Gold Label.


Ah, but the Irish whiskeys would only whet my appetite for what would soon come...

So I set out early on a Saturday morning, hopping a short flight from Dublin to Edinburgh, where I was meeting my wife, Kimberly, and would spend the next 7 days touring Scotland. We both landed at about 9am, and met at our rendezvous point.

Now this trip would be considerably different from my backpacking the previous year. Rather than traveling alone, humping 40 pounds of gear by foot through the Scottish countryside (which is invigorating in its own way), this year I'm accompanied by my lovely wife. Kim is a pretty good sport about the backpacking thing... In fact, I managed to get her to spend 4 of our 14 nights of our honeymoon in 2006 in a tent in Yosemite National Park, including 2 days spent in the back-country dodging bears and bathing in icy streams (Chilnualna Falls, for those familiar with Yosemite).

But we're doing Scotland by car, this time around, which is a dangerous, exciting adventure of a different kind. It starts by piling into a sub-sub-compact rental car and driving on the wrong side of the road. Just to make sure you're paying complete and total attention, 2 short minutes into the adventure, you're testing your nerves against 3 concentric circles of cars swirling about, choosing any one of the 6 different directions their inclination takes them. Yes, the round-about. You see, I only known of one such round-about in all of Austin, and it's in shopping mall parking lot. Luckily, we survived that first one. And got lots and lots of practice with the hundreds of others that we deftly navigated over the next few days.


It took a few hours, but eventually our nerves settled, and driving came more naturally. However, getting into the car was disorienting every single time. I swear, I approached the wrong car door 75% of the time :-)

We drove out of Edinburgh, toward Speyside, crossing the Forth Bridge on the A9--what a marvelous structure! We dipped off the freeway (aka, dual carriageway) as soon as we could, and hit the back roads. Let me tell you, this is the best way to drive across Scotland! We stopped at a couple of small ruined castles by the roadside (Balvaird Castle, for one), and had breakfast and coffee at a little fair. I had been in Ireland for 7 days already, so I had long past the jet lag. This was Kim's first day over, though, so she needed coffee and a big breakfast to reset her circadian rhythm. Refreshed, we headed to St. Andrew's, birthplace of golf.



I'm not much of a golfer, played putt-putt a few times as a kid and been to the driving range one disastrous time as an adult. But St. Andrew's is a site to see. It's absolutely gorgeous, pristine, perfect grass and rolling hills. We kept driving through the heart of the town of St. Andrew's and over to the ruined castle and cathedral. It's a coastal town, with lapping waves, constant wind, and a crystal blue sea. The castle was a fortress itself, overlooking the water, and very near the abbey and cathedral. The original cathedral is well over 1000 years old, and what's left of it is pretty magnificent. This was the center of Scotland church for a few hundred years.


We left St. Andrew's and continued heading up into the Highlands, and the landscape changed entirely. We had left the bustling city of Edinburgh that morning, and saw a bit of the coast at St. Andrew's in the afternoon, and were now just getting into the foothills of the Scottish Highlands. The roads get a bit more narrow and the scenery more spectacular. We past a sign for the Dalwhinnie distillery and pull over for a stop. My bar is generally stocked with the Dalwhinnie 15yr, so I was looking forward to a dram, but alas the tasting room had just closed. Bummer. We took a few pictures and continued our trip up to Aviemore.



Last year, I finished my hike, which started in Buckie, at the Aviemore train station, so I had seen a little bit of the town last year, enough to know that Kim would love it. We parked for a bit, and walked around Aviemore, dipping into a few shops. I have only been to Aviemore in summer time, but it looks very much like a ski town, surrounded by mountains and having sports outfitters on every corner. Next time I hike the Speyside Way, I might just start in Aviemore and buy my gear there, rather than hauling it over on the plane.

Evening approached, and Kim's jet lag started catching up with her, so we made the short drive to Boat-o-Garten, where we had 1 night's reservation at the Boat Hotel. Now Kim had never been there, but it was already famous in her mind. I stopped at the Boat for a short break on the last day of my hike the previous year. I had a cappuccino and an enormous stack of shortbread cookies, heaven in your mouth for a weary hiker. Having taken one bite of one cookie, I thought, hmm, I should take a picture of this :-) The dark wooden table, indirect natural sunlight, and perfect cup of coffee and cookies made for an outstanding shot (if I do say so myself). When I got back to Austin, I had that picture (as well as a couple of others) printed on canvas by Pounds Lab, for wall hangings. So this perfect cup of coffee hangs in our kitchen, and Kim has been dying to have a taste!


So we checked into a nice, sort of Victorian room at the Boat, and headed downstairs for dinner. I'm quite a sucker for Scottish salmon and cask conditioned ales, which I had in one form or another almost every day that week. It was an exhausting day, so we retired quite early, excited about our week of site seeing ahead.

If I've learned one thing having traveled extensively across North America, the Carri bean, and Europe with my wife, it's this... It doesn't matter where we are, or what we're doing. Kim's favorite part of vacation is inevitably the same thing, without fail. And that's breakfast. It's a little hard for me to comprehend, because I don't care much for breakfast. I'll have a bucket of coffee, or a cappuccino, or an ice latte, or some liquid concoction of caffeine. If I've gone for a run earlier in the morning, I might have a little oatmeal or granola to bridge me to lunch. But this gal loves her breakfast. And she really likes it when we stay at a bed-and-breakfast (instead of a tent or Hilton--my two preferences) because she knows that she's going to get her breakfast :-) And on this trip, Kim would have her big breakfast every morning. Not the French or Italian sweet pastry kind. But the full Scottish kind -- two eggs, sausage toast, muffins, cereal, and the freshest milk and butter you can imagine.


She she had her breakfast, and she would need it. This would be the one day that we were going to hike part of the Speyside Way, my favorite part of the hike, and I knew she would love it. The Boat Hotel adjoins with the train station, which runs old fashioned steam locomotives to Aviemore and a few other little towns. After a short wait in the 1920s style waiting room, we took the half hour, slow moving train through the mountains back to Aviemore, with just enough time for tea and shortbread.


It was a perfectly gorgeous Scottish summer day with sunshine and partly cloudy skies. We wandered around Aviemore for a bit, visiting the open air market and a 4000 year old Pictish stone circle, and then made our way onto the Speyside Way for about a 6 mile hike.


As I said above, this was my favorite part of the hike last year. But this time around, it would prove to be even more enjoyable. With no heavy backpack, a short 6 miles to go with only a bottle of water and a camera, and best of all, Kim's company, it was a lovely afternoon. We walked past a hillside that had literally dozens of rabbits popping in and out of holes, practically a scene out of Watership Down. The trail is quiet, the first part under shady canopy of Scottish pines. Eventually, the trail emerges from the woods and out onto a huge, open, heather-laden moor, with 360 degree views of the Cairngorm Mountains. This, too, could be a scene out of the Hobbit. Eventually, we saw the steam locomotive chugging past on its way back to Boat-o-Garten. We hiked lazily, taking lots of pictures, talking about what else we might do on our trip.


Eventually, we made it back to the Boat Hotel, checked out, and sat down in the pub for one more cup of coffee and shortbread. Let me tell you... If you're in Speyside, these are not to be missed! So we packed up the car and headed to a farm a few miles away for what would be one of the highlights of our entire trip!


Let's step back a few years, to Christmas of 2007. Three adorable puppies were abandoned on our neighbor's doorstep one icy night in Austin. Kim and I adopted two of them, and gave the third to a friend-of-a-friend. Kim and I had lightly discussed getting a dog a few times, but never seriously. And suddenly, we were dog owners. Check that, "dogs" owners.


These puppies were barely a few weeks old. We had no idea of their breed, or disposition or anything. Kim's dad is a veterinarian, and he gave them a clean bill of health and identified them simply as "farm dogs". As they grew (and grew, and grew, and grew!), they came to look more and more like Australian Shepherds, or Border Collies. And, of course, we grew to love them :-) So Tiger and Aggie became such an important part of our lives. Now, as some form of sheep dogs, they're extremely smart, and obedient. I have taught them dozens of commands, and they are among the best behaved dogs I've ever seen.


And for these reasons, Kim and I decided to attend a Border Collie Sheep Dog demonstration at a farm in Speyside. Let me start by saying the sheep dog is a remarkable creature. The intelligence and obedience of these dogs are incredible. You've probably seen this video (which borders ridiculousness). But to see a sheep dog in its element, herding, working, and obeying is really awesome. So we spent an hour, with the shepherd introducing us to his 12 working dogs, each by name and a unique whistle that he uses to individually identify the dog to which his command is addressed. He sends these dogs out, one at a time, to herd the sheep scattered across the hillside. He can bark or whistle a command at a dog 1000m away, and the dog will stop dead in its tracks and obey. Wow. A quarter of a mile away, he whistles for the dog to lay down, or slow down, or go right, or left, and it immediately and perfectly complies. I couldn't help but thinking about SSH + Bash :-)


So the dogs reign in the 50 or so sheep scattered across the farm. He also selected one of the sheep for shearing. Kim helped a bit, and described the wool as very "oily". Finally, he showed us how he trains the dogs...using ducks nonetheless. He starts training the dogs at merely 10 weeks old, having them herd ducks, directing them between cones and into a cage. Finally, he brings out the newest litter, 6 puppies barely 3 weeks ago. Adorable :-)


After our official tour was over, we hung around for a bit and talked with the shepherd and his wife. They directed us over the hill where there were a set of "sheep dog trials" currently underway. Thirty-five of the local farmer were engaged in a sort of ancient competition. Each shepherd had one dog, and a huge field below. Three sheep are released at the far end of the field. The shepherd would direct his dog to herd the 3 sheep and drive them through three sets of cones in a sort of "infinity" shape, concluding with driving the sheep into a pin, where the shepherd would separate one and shear it. The competition consists of a set of points and a timer for accomplishing the task. We watched for about a half hour or so, and then needed to make our way to our next hotel.


So we left the farm, and headed toward my favorite hamlet in all of Speyside--Craigellachie. I spent two nights the previous year at the Highlander Inn in Craigellachie, which were my two favorite nights of my hike. I couldn't wait to get back, and particularly to share it with Kim this time. Craigellachie is absolutely in the heart of the single malt Scotch Whisky country. The Highlander Inn is "the perfect hotel in the world", as far as I'm concerned.


It has 5 comfortable rooms above the pub, free wireless Internet, and views of the mountains over the Speyside River. If that's not enough, you can sample over 500 bottles of Scotch Whisky, or 6 different cask conditioned hand drawn Scottish ales. The food is excellent (of course the salmon, but also try the haggis-stuffed chicken). But the best part is absolutely the company. I spent a good 10 hours in the pub last year, cavorting with the locals, telling stories of Texas and Louisiana, and listening to their stories of the Highlands and Speyside. Duncan, the proprietor, is an outstanding gentleman. And without surprise, I ran into many of the same people I met the previous year in the very pub, including my friend Roy (local tour guide, see aboutspeyside.com). Kim and I sat at the bar for several hours, enjoying the accents and the conversations of the vivacious locales. I took quite a bit of advice on their favorite whiskies, sampling drams of Mortach 16yr and Ben Riach 16yr between each pint of ale.

Kim sprung out of bed the next morning, ready for her next round of breakfast in the charming dining area on the second floor. Duncan answered some logistics questions for us on our touring options, and in fact, called in a favor with the Aberlour Distillery and booked us on a 10am tour. The Aberlour 12yr is another of my stock single malts--double barrel aged in port wood, smooth, almost sweet finish. Kim isn't much of a hard liquor drinker, but she loves a good tour. We often visit the 20-or-so wineries in the Hill Country near Austin, and love to do so in Napa, Sonoma, and Italy too. So she kept an open mind, and thought that if she learned more about it, she'd enjoy it more. The tour was very thorough, and informative. We saw all stages of the whisky making process, and learned the story behind the Aberlour Abundah, and St. Drosdan's well. The tour completed with a tasting of 6 different whiskey's (well, 5 really--one was just raw spirit straight off the still--yuck!). And, alas, Kim still hated the taste of whisky :-) Oh, well... I have my designated driver, then.


So we walked a few miles down a trail behind the distillery, and eventually made our way back to the car. Kim, breakfast. Dustin, whisky. Kim's turn... So we headed to Ballindalloch Castle. Ballindalloch Castle is a still-lived-in monstrosity of a castle, owned by the family that developed Aberdeen Angus beef (which is quite tasty, by the way). We toured this lavish castle, seeing some old, ornate furniture, lots of gold, crystal, and tapestries that wouldn't even fit in our house. The watchtower was my favorite part, which included a hole that guards could use to dump sewage on invaders. Take that! We had lunch at the cafe near the castle, where I had an Aberdeen Angus roast beef sandwich. Yum. We also stopped at another ancient stone circle, which is in the middle of a pasture.


Next, we drove way up the mountain to the hilltop town of Tomintoul (another excellent single malt). Just outside of town is an abandoned quarry, with a little path to the top and spectacular views of the valley. We got some really nice pictures. From there, we drove through the Glenlivet Estate (which is owned by the Crown). We stopped in a little town hall where we had some coffee and Kim perused the homemade crafts for sale. Also nearby is an ancient bridge dating from the 9th century. We drove by the Glenlivet distillery, but since I had toured it last year, we skipped it.


There are plenty of pastures around, most of which have Highland cattle, which look something like a hippie version of the Texas longhorn!


We made our way to Dufftown next, which calls itself the "Whisky Capital of the World". Here we visited Mortlach Kirk, the 3rd oldest church in Scotland, founded in 566. In the cemetery, you can see the ancient Pictish standing stone, called the Battlestone. Legend has it that King Duncan (as in Macbeth) prayed here before defeating the Viking invaders. There are graves in the cemetery that are over a 1000 years old. Creepily enough, many of them didn't use coffins, or the wooden coffins have deteriorated, and if you look closely, there are bone and teeth fragments through the gravel you walk upon!


We also stopped at Drumin Castle, another spectacular ruin. After that, we drove past the Glenfiddich distillery, and walked around the ruined Balvienie castle. As it was getting late, we headed back to Craigellachie for a pint in the bar, and some of that golden conversation with the locals.

So this particular day of the vacation was special for me, as it was my birthday. I turned 30 years old. It was a wonderful way to spend it, though. Soaking in the Scottish culture, indulging the senses, enjoying the vacation of a lifetime with my adorable wife. We capped the day off with a five-star meal at the Cragellachie Inn. It, too, has remarkable selection of 600+ Whiskies on the wall. As an aperitif, I had a 30yr Tamdhu. And for dessert, the most remarkable whisky I've ever tasted--a 40yr Strathisla. Wow. Buttery, creamy, perfectly smooth, and indulgent. Kim bought me a bottle as a belated birthday present once we got back home. We called it a night, and made our way back to the Highlander (literally across the street).


Up until this point, we had reservations booked in advance for all of our accommodation. However, the next two nights were totally open. We could really go where the wind took us, which is a really cool feeling. After breakfast (of course), we took a quick look at Thomas Telaford's gorgeous Craigellachie Bridge. We then made the short drive over to the Speyside Cooperage, where we took an education tour of the process by which whisky barrels are made (or reconditioned). Really, interesting, sort of a different kind of a tour. We both enjoyed it very much.


So we took a look at a map and then made our plans for the next two days. We decided to drive to the Isle of Skye, and see the west coast of Scotland. So we drove up toward Inverness, stopping for a dram at the Macallan distillery (outstanding, the Rolls Royce of whisky, if you have any respect for Scotch, you must have at least a bottle of Macallan 12yr on your shelf), and passed by the Strathisla distillery near Rothes.


We passed through Elgin, and Inverness, and made our way to Loch Ness. We drove the full length of Loch Ness (didn't see Nessie but saw plenty of Nessie-related crap for sale). We did stop for a long tour around Urquhart Castle. This castle must have been one of the most incredible castles/forts of Scottish history. It's on a prominent point, guarding the length of Loch Ness. An entire city flourished within its walls, and it protected inner Scotland from numerous Viking invasions. I'd highly recommend this castle to drivers by.


From here, we drove along some very small roads, winding between mountains and lakes, making our way all the way over to the east coast to the islands. The mountains were considerable larger than those in Speyside, and the landscape far more stark. Heavy fog, and clouds rolled all around us (compared to the relatively sunny Speyside area). Of far more concern, though, was the relatively few B&B's we saw with vacancy. Oh, don't get me wrong, there were plenty of B&B's. But most were sporting their "No Vacancy" signs. Kim was getting more than a little worried. We were still a long way from Skye, and were starting to wonder if we'd even be able to find a place at all, without reservations. And while I have no problem sleeping in the car (or on the ground, for that matter), that isn't exactly Kim's cup of tea. So we stopped at the next B&B that had vacancy. The price reflected the quality. Merely 30 pounds, Kim went in to check it out. Yikes. Trinkets everywhere, and the proprietor gave off the "creepy old man" vibe. We had no interest in the Bates Motel. We also stopped at a ranch which had vacancy. I like the idea of a horseback ride through the mountains. But the only accommodation they had was bunk beds, dormitory style, shared with others. Again, not exactly Kim's bag.


We kept driving, past the site of the final stand of the Jacobite Rebellion (and the last shots fired on mainland Britain). In the worst case, if we couldn't get a room near Skye, we would just skip it and drive to Fort William, where there seemed to be much more vacancy. We ended up staying at a *picture perfect* B&B just past Glen Shiel in Inverinate. A cute, sweet little old lady with a gorgeous cottage on the loch, for merely 50 pounds/night. We went to a pub (The Claggan) that night and hung out with the locals, including a very friendly chap named Ian who worked at the Eilean Donan Castle. We saw him the next morning and he actually let us in for free! We drove all over Skye with the rest of our day, stopping at the Talisker distillery for a tasting, and touring Dunvegan Castle. We found a couple of neat, old cemeteries and a standing stone. The roads are absolutely tiny, mostly single lane. When you encounter another car, you have to pull the side, and only one goes past at a time. The mountains and sea are spectacular.


We spent two nights at the very cozy bed and breakfast. As usual, the breakfast was outstanding, but more impressive was the view of the glassy lake in the front yard. This place isn't far from heaven ;-)


Even though Skye was rather touristy, we had a great time. Next time, though, we'll probably head a bit further North, toward Ullapool and some of the less traveled islands. We drove back through Fort William, where we stopped at the Ben Nevis distillery for a taste (probably the only one that I haven't been impressed with). As usual, Ben Nevis itself was shrouded in clouds and fog, so we couldn't even really see it. We hit a couple more ruined castle roadside stops. Oh, and also Rob Roy's grave was prety cool, in a beautiful little cemetery. We drove through Stirling, and could see the castle on the top of the hill, but we didn't.


We did, however, stop at the Falkirk Wheel, which amazing beyond words! I said that there were two real highlights of our trip--the sheep dog demonstration, and absolutely the Falkirk Wheel. Simply breathtaking. I watched it for nearly an hour. This machine simultaneously lifts and lowers a boat the equivalent of 13 traditional locks, without loosing a drop of water, and is powered by a series of small electric motors for less than 10 pounds of electricity per day! This thing is a marvel of the modern world. If you're an engineer living on the isle of Britain and you haven't been to the Falkirk Wheel yet, you should make plans to do so immediately!


After Falkirk, we made our way back to Edinburgh, where we'd spend our final two nights at the Edinburgh Hilton. Little did we know, we had just immersed ourselves in the middle of the Fringe Festival. We caught several comedy shows and theatrical performances--all for free. That was a nice surprise, as we weren't expecting that. The Rowan Caves are really neat, which is where we spent most of our time. We also toured the Edinburgh Castle, though the weather was very, very poor (rained the entire time).


And that's about it for our trip across Scotland! We drove several hundred miles, covering so much of Speyside, the Highlands, Skye, and Edinburgh. And already I have the itch to go back--maybe winter next time. A snowy Speyside, perhaps? Some skiing in the Cairngorms? Ah, that sounds nice!

:-Dustin

September 29, 2009

What's the smallest laptop with VT on the market?

Howdy all-

In the course of the Ubuntu Server Team's development of the Ubuntu Enterprise Cloud and Eucalyptus, we'd like to travel with a prototype "cloud" in our luggage.

What's the smallest laptop you know about that has Virtualization Technology (VT) extensions on the CPU?

I currently have a couple of 12" Thinkpads (x200 and x61) with VT. Looks like the Dell Vostro 1220 is another 12" with VT.

Have you spotted anything smaller than 12"? I'm only interested in laptops/netbooks (must have built in keyboard/video), it must have VT, and it must be smaller than 12". Suggestions?

:-Dustin

September 26, 2009

Sep 20 – Sep 25 Wrap-up


Spent most of my week in Portland to attend conferences.

Conferences

  • Attended LDAPCon 2009 and published report.
  • Attended LinuxCon 2009.

Image Store Proxy

  • Updated image-store-proxy to 1.0. This version brings support for gpg signed images. Still need testing against the real-world Canonical Image Store infrastructure.

A summary of LDAPCon 2009


On Sunday, September 20th and Monday, September 21st I attended LDAPCon 2009 in Portland, OR. Most of the open source projects were there – with the notable absence of Port 389 (Redhat) – as well as some vendors (Apple and UnboundID). Most of the slides are available online.

Apache Directory project

The Apache Directory folks gave several presentations:

Apache Directory Server provides an integrated product with most of the standard network services: in addition to ldap, dns, dhcp, ntp and kerberos services can be enabled as part of a deployment. Kerberos support seems to be in early stage as it almost works. Another interesting aspect of the project is its integration with the Eclipse environment. Apache Directory Server is embedded in Apache Directory Studio. The latter provides a management tool for Directory administrator. If the Eclipse integration in Ubuntu is improved Apache Directory Studio would be a very good addition to the archive.

An overview of implementing replication in the Apache Directory Server project was given. RFC 4533 is used as the basis for LDAP replication in OpenLDAP. The goal here was to be able to replicate between Apache Directory Server and OpenLDAP. This may be the start to a standard replication protocol between different directory products.

Three components needed to be implemented:

  • the consumer part is the easiest and can be a standalone component. It receives LDAP entries updates and can do whatever it wants with them. It reminds me of similar requests I heard at the MySQL User Conference last April where people were interested in having an easier access to the MySQL replication log.
  • the producer is more complex to implement as it requires to keep a log of the modifications done on the server.
  • conflict resolution is the hardest part and mandatory if multi-master is to be supported. The Apache Directory Server decided to implement a strategy of last writer wins as they’re trying to not require any user intervention for conflict resolution. I’m not convinced this is the best strategy though.

While implementing replication support they’ve also added support for store procedures and triggers.

LSC Project: LDAP Synchronization Connector

Corporate environments usually have multiple identity repositories and keeping all of them in sync can be quite a challenge. The LSC project aims at automating the task of keeping all identity stores up-to-date. Written in java it can read and write to any database or LDAP directory. On-the-fly transformation of data sources are possible and the framework tries to make it easy to implement new synchronisation policies.

Another great tool that could be added to the directory administrator toolbox to help integrate Ubuntu in existing infrastructures.

Storing LDAP Data in MySQL Cluster (OpenLDAP and OpenDS)

This was a joined presentation between the OpenLDAP and OpenDS projects. A new backend has been added to store entries using the MySQL Cluster NDB API. The main advantage is to be able to access the same data over SQL and LDAP as well as providing a highly-available infrastructure with data distributed on multiple nodes. Both OpenDS and OpenLDAP have worked together to create a common data model highlighting that cooperation does happen in the LDAP space.

A Panel discussion among the representatives of the various LDAP Projects on roadmaps

Sunday ended up with a panel where representatives of different directory vendors answered questions from the audience. Each open source project briefly outlined a few points they were trying to improve: documentation for OpenLDAP, data migration for Apache Directory and multiple schema support for OpenDS. The issue of virtual directories was also discussed with the need of more GUIs to cover administration tools as well as workflows. Apache Directory Studio was given as a potential good starting point to build these higher level tools. The subject of standard ACL’s was also covered. It seems that this is still a sensitive issue in the community and projects are still arguing about a common solution. One option put forward was to look at the X500 ACL model and start from there.

The last item of discussion covered how to expand the user base of directories. The world of directories is rather small and its use cases are usually associated with Identity Management (User and Group, Authentication). Having good client APIs was mentioned as an option. However the whole group ran out of ideas quickly and got kind of stuck in front of this problem.

Directory Standardization Status

Directory standardization happens within two bodies: X500 in ISO/IEC and LDAP in IETF. The most important topic currently discussed in both bodies is password policies. A new draft of an IETF document is being worked on by Howard Chu and Ludovic Poitou.

Other topics being worked on cover:
  • Internationalization (with Unicode support in LDAPprep and SASLprep)
  • simple LDAP Transactions (to cover adding entries to different containers)
  • replacing DIGEST-MD5 with SCRAM
  • vCard support

On the front of Directory Application schemas support for NFSv4 Federated Filesystem and an Information Model for Kerberos are currently being worked on with drafts available for review.

The question of starting a new LDAP working group within the IETF was raised. Topics that could be covered include:
  • LDAP Chaining Operation
  • Access controls: based on the X.500 model with extensibility added.
  • LDIF update
  • LDAP Sync/ LDAP Sync-based Replication
  • Complex Transactions
  • Password Policies
  • Directory views
  • Schema versioning

LDAP in the java world

LDAP support in java is being actively worked on especially on the SDK front. OpenDS, Apache Directory Server and UnboundID have released new open-sourced SDKs to improve the aging JNDI and Netscape java SDKs. All of them are rather low-level implementations. The three projects are also working together to find a common ground.

There is also some progress made at the persistence level. The DataNucleus project gave an overview of adding LDAP support to the standard JDO interface. The goal is to provide a reference implementation of JDO for an LDAP data store.

Unified Authentication Service in OpenLDAP

Howard Chu gave an overview of the new modules developed in OpenLDAP related user authentication. Based on the work from nss-ldapd the nssov overlay provides integration with the pam stack as well as the nss stack. Disconnected mode in the pcache overlay has been added in the latest version of openldap as discussed during the Ubuntu Developer Summit last May. Most of this work is already available in Ubuntu Karmic and improvements should be made during the Lucid release cycle.

Another interesting module is the integrated certification authority. If a search request for the userCertificate and userKey attributes for an entry is made and these attributes don’t exist they’re generated on the fly. This should help out in creating an X.509 base PKI.

LDAP Innovations in the OpenDS project

The last session of the conference was given by Ludovic Poitou of the OpenDS project. New features available in OpenDS include tasks as well as extended syntax rules. Time matching rules have also been added so that queries like “give me entries that have a last login time older than 3 weeks” can be expressed directly in ldap and processed by the server. That brought some interesting issues when clients and servers don’t share the same timezone.

A few gems from beer conversations

After the official sessions ended most of the attendees congregated to have diner followed by beers. Howard showcased his G1 phone running slapd while Ludovic was showing off an LDAP client application on his iPhone. And of course by then end of the conference both systems were connected: the iPhone was able to look up contact information on the G1 running slapd.

On an unrelated note OpenLDAP is faster than OpenDS, even in beer drinking. However the OpenLDAP project was compared to a Beetle car with a Porsche engine whereas OpenDS was actually building a Porsche.

Even though not all the players in the directory space were represented at the conference, most of the key players from the open source world were there presenting their work. Friendly competition exists amongst the different projects which turns into cooperation on topics that matters such as interoperability and data formats.

It seems that the directory world is rather small and its use cases are restricted to specific situations compared to RDBMS. This is rather unfortunate as directories offer a compelling alternatives to databases as a data store infrastructure. The community seems to be aware of this issue and is looking into breaking out of its traditional fields of applications.

September 23, 2009

Server Team 20090922 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

  • soren to add manifest files to UEC images build system for alpha6: Done
  • Daviey to update Asterisk 1.6 to RC1: In bzr and PPA, need sponsorship

ACTION: Daviey to get his Asterisk 1.6RC2 update sponsored

  • soren to publish ec2-version-query in a more appropriate place: assumed to be in hands of release team

ACTION: soren to clear out status for ec2-version-query publication

  • soren to automate updating of ec2-version-query”: Blocked by previous ACTION
  • zul to ensure rabbitmq-server gets reviewed and promoted: Done
  • soren to sponsor the patch for bug 420581 and update his vmbuilder on nectarine: Done
  • ivoks to file FFe for the redhat-cluster update: Done, bug 429834

Carried-on items, for reference:

ACTION: soren to automate updating of ec2-version-query” (once publication is resolved)

ACTION: kirkland to open discussion on how to best solve the remaining configuration options on Moodle appliance

ACTION: kirkland to get help from soren and smoser on proper UEC-compatible image generation

ACTION: kirkland to discuss with niemeyer and nurmi about image store integration testing

Alpha6 postmortem analysis

Alpha6 went out last week and was globally a smoother release process than Alpha5 release. A few todo items have been identified, to be implemented for beta release:

  • Define seeds for vmbuilder to use (cjwatson, soren)
  • MIR all non-main packages used in images (smoser), in progress

This includes using euca2ools, which in the latest version is ec2-interface compatible.

ACTION: soren to update to latest euca2ools

  • Publish ec2-version-query in a appropriate place (soren)
  • Automate image publishing (smoser)
  • Add build toolchain version numbers to manifests (soren)

ACTION: soren to add image-generation-toolchain version numbers to manifests

Roadmap review: UEC/EC2 images bugs

The karmic kernel used in the Alpha6 images turned out very good. The only issue reported about it is a few missing config options (bug 428692) and lack of modules (bug 429169).

The most blocking issues in those lists are the MIR bugs since they don’t strictly depend on the team and might raise last-minute extra work.

ACTION: zul to follow up on the UEC/EC2 packages MIR status

A bug needs to be filed about the fact that the images include unsupported packages, and targeted to beta.

ACTION: smoser to file one bug for the fact that the images include unsupported packages

Roadmap review: Packaging and integration of Eucalyptus 1.6

ttx performed some feature/usability testing on the Cluster install side that uncovered multiple issues preventing autoregistering of components to work properly. The blocking issue is that it is no longer possible to register using “localhost”, the external IP addresses must be used instead (as they will be passed to other components), see bug 434651. This might involve asking the user for the cluster’s user *and* node facing IPs. euca_conf should also support for forcing local copy of keys (bug 434651).

erichammond asked about euca2ools having to provide ec2-* links to the euca-* commands. At this point this is still under discussion. That discussion needs to happen somewhere, so a bug will be filed to support it.

ACTION: ttx to file bug about providing ec2-* command names which call euca2ools

Roadmap review: Virtual appliance

kirkland reported a slowdown in appliance creation last week, as other priorities kicked in. He encountered some issues creating the images, mostly related to vmbuilder and debconf questions. Several community members contacted him to propose their own VM appliances and appliance builders. mdz said that the reference appliance must be produced this week. We need a fully reproducible build system but producing a reference appliance to exercise UEC and the image store is more urgent. mdz will take the opportunity of being at LinuxCon with kirkland to discuss this more.

ACTION: mdz to sync with kirkland on Virtual appliance status

On the image store side things seem to be under control, with niemeyer pushing last fixes and mathiaz packaging them.

Assigned and to-be-assigned bugs

Review of the bug list raised the following remarks:

  • cjwatson considers himself almost entirely done with the eucalyptus tasks that were assigned to him
  • nurmi mentioned potential need for a few more bits in the init script to take into account /etc/eucalyptus/installer-cc.conf and will follow up with filing a bug if necessary
  • soren has a lot of bugs assigned, ttx will reassign a few to himself
  • mdz noticed that wishlist bugs should be omitted from that list unless they’re targeted

ACTION: ttx to poke QA team about omitting untargeted wishlist bugs from the buglist and add something to those pages which tell you who to contact about them

Weekly SRU review

Most bugs in the recently-fixed bug list pertain to karmic only, so nothing stands out as SRU-worthy. No nominations this week.

On the list of accepted bugs with an assignee, zul said most of them are now waiting to be accepted into *-proposed. mathiaz said we should start using bzr branches to handle the SRU process, and wanted to involve sbeattie in the process.

ACTION: mathiaz to involve sbeattie in the Weekly SRU review process

Open Discussion

erichammond said he still can’t update importance on any ec2-images bugs, as he is not a member of bugcontrol yet. mdz fixed it.

Agree on next meeting date and time

Next meeting will be on Tuesday, September 29th at 15:00 UTC in #ubuntu-meeting.

Soren should soon start a discussion about moving the meeting to another time, as the extended duration of the meeting doesn’t fit “people in CET with families” constraints.

September 20, 2009

Atlanta LinuxFest the Day After

Wow, I'm still in a daze. Atlanta LinuxFest (ALF) was an incredible event. It was held at the IBM facility in Atlanta on Saturday.

Some background first. My wife Amber (akgraner) was on the planning team, so I was watching this event come together from the periphery. I would hear her on planning calls and I never really understood what it takes to put an event like this together. My hats is off to all involved, I can say it gave me a new appreciation for all the work that goes into community events like this.

This was ALF's 2nd year and as such was expected to be small. It was far from that, they had over 600 registrations and at any given time during the event there were 300+ people at the event. This event was planned by 4 people with limited resources and they pulled off one hell of a LinuxFest.

The speaker line up was diverse and impressive, there was something for everyone. Additionally Amber had organized an UbuCon and it was a big hit. They had sessions topics like "Community Leadership" & "Burnout", every time I went by the UbuCon area it was packed and they had some intense discussions going on.

The Ubuntu Kernel Team took advantage of the event to get some Karmic testing done on the plethora of laptops/note/netbooks that were there. Manjo Iyer from the Kernel Team ran the testing. I don't have the exact numbers of what makes and models were tested (we still have to sift thru the data), I do believe it was well over 100 people and some had 3 & 4 machines with them. Lots of bugs were filed. I want to put out a heartfelt "Thank You" for everyone that came by for testing. You guys will make Karmic a far better release.

I gave a talk on Ubuntu, Canonical & the Ubuntu Kernel Team. I had given a talk at the Southeast LinuxFest back in June and received lots of feed back about the content. I was surprised to learn that people wanted to know about Canonical, a bit about it's structure and how Ubuntu fits in. So I incorporated lots of that back into the presentation. I also stressed how the Kernel Team is looking to expand its community, and if you want to participate you don't have to be a kernel developer, or even a programmer! We welcome anyone who is willing to test, triage or help us organize. Like in any community we need people with diverse skill sets.

Steve Conklin of the Ubuntu Kernel Team gave a talk called "Debugging the Kernel". This talk originated from another Kernel Team member Colin King (cking). The talk is basically a collection of all the wild & useful debugging techniques that Colin has come up with over the last few years.

John Johansen & Stefan Bader from the Ubuntu Kernel Team gave a rehash of Greg Kroah-Hartman's "Write a real working Linux driver" tutorial. The tutorial consisted of a Ubuntu USB live stick tricked out with compilers, headers and git tree. Users would boot the live stick so that they would have a consistent development environment. John then walked them through the basics of git, kernel device drivers and in the end the users wrote a device driver that would work with a GoTemp USB Thermometer. They had 16 thermometer devices and in the end the temperature could be read by reading a file in the /sys file system. Each session was full, I just wish we had more devices so that everyone would have had the chance to fully participate, not just watch.

Dan Chen of Ubuntu Audio fame gave a great talk on debugging audio. Judging by the size of the crowd in his talk, audio is still an issue with quite a few users.

Suse, Red Hat & Fedora were all there with booths and talks as well, however I would say that the mind share at the event went to Ubutnu. You couldn't turn around without hearing the Ubuntu login music, seeing Ubuntu stickers, banners, t-shirts everywhere!!!!

Surprisingly quite a few folks have been running Karmic in some state of Alpha for quite a while!

The day started with a Video Podcast from Mark Shuttleworth specifically for the event. Mark was about to announce the name of 10.04 when they cut the video with a slide that said "Find out at the UbuCon!"... dooh! It left everyone hanging for about another hour.

In the UbuCon area they had monitor set up with people crowed all around waiting for the announcement. They played the whole video from start to finish and finally after much anticipation, Mark announced that 10.04 would be called Lucid Lynx. This was quite a departure from other "naming announcements" where he would send out an email or post it on his blog. It was really special that it was announced at a UbuCon at a community event!

There is so much more, but I'll leave that for the other blogers... I have to catch a plane on my way to LinuxCon & Linux Plumbers in Portland!

~pete

September 19, 2009

Sep 11 – Sep 18 Wrap-up


Image-store-proxy

Package image-store-proxy to enable the Image Store tab in Eucalyptus. The package (python-image-store-proxy) has made its way to main and on the -server isos in time for alpha6 with the help of Thierry and Kees.

Server-karmic-directory-enabled-user-login

Kept on investigating the use of puppet to build an ldap/krb5 infrastructure on EC2. Integrated dnsmasq and puppetmaster configuration. Discovered a few bugs along the way and reported them upstream. My current work is available from an LP branch. And puppet is awesome!

Alpha6 ISO testing

Loads of alpha6 testing.

Landscape-client Stable Release Update

Reviewed the landscape-client and smart SRU requests from the Landscape team.

Bug scripts

With the help of Brian my bug scripts are now regularly run on qa.ubuntu.com. All bug lists used in the SRU review and the triaging process can be found on qa.ubuntu.com.

Misc

Updated my status report script to publish a draft of my activity report on my blog as the weekly “wrap-up”.

September 16, 2009

Server Team 20090915 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Review ACTION points from previous meeting

  • smoser to tag existing UEC image bugs with “uec-images”: Done
  • soren to ensure that smoser can update the UEC publishing scripts: Done
  • smoser to add MD5SUMs for UEC images: Done
  • soren to add manifest files for UEC images: Code landed in VMBuilder, will be added to build system today

ACTION: soren to add manifest files to UEC images build system for alpha6

  • smoser to open dialog with IS about automated publishing to EC2 and agree on a plan: Done, ticket 35660
  • soren to automate updating of ec2-version-query: Blocked on automated publishing

On that subject, nijaba mentioned it would probably be good to have a human-readable equivalent page updated at the same time. This should be discussed on ML and/or as a future meeting egenda item.

  • soren to publish ec2-version-query in a more appropriate place: soren needs to get back to slangasek on that.
  • smoser to add ec2-images tag to the relevant bugs: Done
  • nijaba to fold #ubuntu-ec2 and #ubuntu-cloud into #ubuntu-server: Done
  • soren to triage all eucalyptus bugs, and use the ‘eucalyptus’ tag for bugs which should be escalated to the eucalyptus team: Was done, but now needs a refresh
  • kirkland to build a proof of concept alfresco appliance: This was retargeted, see below.
  • mathiaz to get niemeyer’s proxy code packaged: Done, MIR ready.
  • zul to ensure rabbitmq-server gets reviewed and promoted: Still waiting on MIR team, zul will ping them again today.
  • mathiaz to upload openldap 2.4.18: Done
  • kirkland to speak with marjo about how to get qemu-kvm tested prior to release (and more generally server applications like it): Done, will hold a bugday in 2 weeks
  • mathiaz to get a server dev team set up in LP and work with TB to get it set up for archive reorg: This is believed not to be needed for now. However mathiaz will have a look at the package sets again.
  • ttx to update server team Roadmap to reflect current projects: Done
  • Daviey to call for testing of Asterisk 1.6: Done

On that subject, Daviey mentioned the opportunity to update to RC1 (we are using 1.6beta4 now and are expected to move to 1.6 release in the end). ttx advised to release early and often, especially since RC1 is a bugfix-only update from beta4.

ACTION: Daviey to update Asterisk 1.6 to RC1

For reference, here is the list of ACTIONs from last meeting still in progress:

ACTION: soren to automate updating of ec2-version-query

ACTION: soren to publish ec2-version-query in a more appropriate place

ACTION: zul to ensure rabbitmq-server gets reviewed and promoted

Alpha6 remaining actions

The alpha6-milestoned buglist is empty of server-related items, so the question was up if any alpha6 targets were missing from the list. Bug 413789 was mentioned, but retargeted to 9.10-beta, pending help from mvo on resolution.

Some release process improvements were also targeted for alpha6, here is the status on them:

  • Add signed MD5SUMS (slangasek): slangasek has now access to nectarine and proceeded on adding that.
  • Add manifest file for each image (soren): 87% done, see ACTION above
  • Automate publishing of AMIs to EC2 (smoser): Deferred to beta release
  • Automate updating ec2-version-query (soren): Blocked on automatic publication, so deferred
  • Publish ec2-version-query in a more appropriate place (soren): needs more discussion with slangasek
  • Ensure inclusion of relevant news in release notes (erichammond): Blocked on manifest generation

Roadmap review: UEC/EC2 images bugs

smoser compiled a detailed status at http://paste.ubuntu.com/271503/. ttx asked about bug 420581 which is targeted for alpha6: it needs soren to pull the patch, review it, sponsor it, and then update his vmbuilder on nectarine, which should be completed for today.

ACTION: soren to sponsor the patch for bug 420581 and update his vmbuilder on nectarine

smoser reported last week release of ami-a40fefcd and ami-3fb25256 to ubuntu-ec2, ubuntu-cloud last week. This is the first image we’ve published with karmic kernel by default. We’ve got fairly good feedback, the only real issue raised is on bug 428692. smoser is also working with jono, jorge and ara to get a community test plan together. ttx asked about bug 418130, which should be targeted for alpha6 since we are to release it with a Karmic kernel, and smoser retargeted it. erichammond asked whether kernel modules and a “loop” fix were going into Alpha6. smoser answered that those are 9.10-beta targets, though the kernel modules fix might make it into alpha6.

Roadmap review: Packaging and integration of Eucalyptus 1.6

As already mentioned, the list needs a refresh. No other comments from the crowd.

Roadmap review: Virtual appliance

niemeyer was not present to the meeting, but the image-store-proxy is currently blocked on the image-store-proxy MIR. During the ACTION review, kirkland mentioned a change in reference appliance. Alfresco was abandoned due to its dependency on sun-jdk, partner-only packaging and lack of community testing. The current target is Moodle, which is in main, all open source and has a nice first-time setup via a web front end. A proof of concept image was produced at http://rookery.canonical.com/~kirkland/ubuntu-9.04-moodle.qcow2.bz2 (260Mb). One remaining issue is the need to dpkg-reconfigure on first login, forcing the user to make a few selections. Most of them can have sane defaults, but FQDN probably can’t. Various possible workarounds were suggested, but none of them stood out as appropriate. This should be further discussed off-meeting.

ACTION: kirkland to open discussion on how to best solve the remaining configuration options on Moodle appliance

The appliance as generated now cannot be run in UEC (qcow2 format), nor would it be able to show up in the Eucalyptus Image Store.

ACTION: kirkland to get help from soren and smoser on proper UEC-compatible image generation

ACTION: kirkland to discuss with niemeyer and nurmi about ImageStore integration testing

Roadmap review: Other specs

On the Cluster stack front, ivoks reported everything for pacemaker as done and in karmic, ready for shipping. On the other hand, RHCS needs one more update (bug 429834). Looking at the extent of the update, this certainly requires a FeatureFreeze exception.

ACTION: ivoks to file FFe for the redhat-cluster update

With upload of OpenLDAP 2.4.18, the directory-enabled-user-login spec is feature-complete. It now needs testing.

Assigned and to-be-assigned bugs

The buglist at http://qa.ubuntu.com/reports/team-assigned/canonical-server-assigned-bug-tasks.html doesn’t show any bug needing to be reassigned. ttx wanted some feedback on proposed solution for bug 425928, and got a general agreement on that.

Weekly SRU review

The proposed bugs lists were reviewed and appropriate nominations were accepted. The list of accepted bugs with an assignee was also reviewed and updated.

Agree on next meeting date and time

Next meeting will be on Tuesday, September 22nd at 15:00 UTC in #ubuntu-meeting.

September 15, 2009

Byobu on a Palm Pre

Having owned Palm Treos 600, 650, and 755p, I finally bought a used Palm Pre off of Craigs List earlier today. And although I've had the phone less than 24 hours, I already love it!

Following a couple of guides online, I obtained root on the device without a problem. This is where the real fun begins...

I was absolutely delighted to find that GNU screen is installed by default on the device! Which, of course, means that byobu should run too ;-)

I had to fix a few things in byobu, since I had never tested it on ARM, or inside of busybox, but it's working quite well! See this screenshot:


So within the first 12 hours of owning this little miracle, I've managed to:
  • sync all of my contacts and calendars from Gmail and Facebook
  • obtain a root shell
  • run screen
  • install and run byobu (using byobu-export from the unreleased byobu-2.33)
  • install and run an ssh server
  • ssh to the device
  • tether (working perfectly with Ubuntu, writing this blogpost tethered right now)
  • actually lower my monthly service plan with Sprint
  • sign no contract since I bought used via CraigsList
But the best part... Real Linux, with a root shell, running on my phone. I've been waiting for this day for over 10 years. I'm choking back tears. Wow.

:-Dustin

September 12, 2009

uninstall sun-java6

With the vrms meme raging on Planet Ubuntu, I noticed some people still have sun-java6 installed. I’ve been using openjdk-6 since Hardy, and everything I use works fine with it (e.g. Vuze, Catan, Eclipse, FreeMind, and even Facebook’s photo uploader thing). Given the Ubuntu Tech Board’s “remove sun-java6 from the archive” Agenda item, it’s possible sun-java6 (being redundant and orphaned) would go away in Ubuntu Karmic (sun-java5 is already gone because Sun will drop support for it in October, and sun-java6 is not far behind).

I’ve heard some noises about openjdk-6 not working for people, but I haven’t actually seen any direct technical evidence of something working in one and not the other. Seeing as Jaunty’s OpenJDK was certified by Sun, it would be very interesting to find failures, as this would indicate that Sun’s certifications are missing something.

Looking for examples of failures in Google, I couldn’t find anything with obvious test-cases that failed with openjdk-6. I suspect I’m just not trying hard enough, but I’m curious what other people have run into.

Test for yourself:

sudo apt-get purge sun-java6-jdk sun-java6-jre sun-java6-bin sun-java6-plugin
sudo apt-get install openjdk-6-jdk icedtea6-plugin

(Don’t forget to restart your browser.)

September 11, 2009

Sep 07 – Sep 11 Wrap-up


Server-karmic-directory-enabled-user-login

Upload new sssd package to fix lintian errors and pull two fixes from upstream. Brainstorm with upstream about testing the package.

Prepare and upload openldap 2.4.18 to Karmic once the FFe was granted. That should complete the last part of the specification and brings disconnected mode support on the client via the cache overlay.

Looked into using puppet to build an openldap/krb infrastructure to test all the directory related components on the client side (sssd, openldap pcache overlay). The idea is to be able to pull up and down complete environments within minutes using a combination of EC2/UEC and puppet.

Follow up on puppet promotion into main for karmic.

Ended up writing a custom puppet type to handle slapd modules using the default karmic configuration. This gave me a good overview of how puppet is working.

Imate-store packaging

Looked at packaging. Follow-up call with Gustavo. Should have a package ready on Monday in time for alpha6. More polishing will be done for beta.

Apport in the default server install

Add apport to the default server install as requested by steve beattie for the karmic-qa-apport-in-ubuntu-server specification.

Linux-virtual missing virtio modules

Chase down and confirmed that linux-virtual kernel doesn’t have any of the virtio modules. Bug 423426 is milestoned and should be on the release team radar. This has a high importance as virtio vms cannot boot in karmic. Tim is working on it.

Mysql maintenance

Caught up on (lots of) mysql 5.0 and 5.1 bugs. Updated DebuggingMysql page in the process of triaging bugs.

Upload mysql 5.0 and 5.1 to fix a couple of bugs. Both mysql-server-core-5.{0,1} packages provide mysql-server-core which should be used by packages requiring the mysqld binary (such as akonadi).

Sru-workflow

Write up a script to get a list of ubuntu-server SRU bugs assigned to people. This produces the remaining list to be reviewed during the team meeting with the updated SRU workflow in the ubuntu-server team.

Sponsoring

Reviewed checkbox merge proposal from Marc. Asked for a FFe as there is one new feature.

LDAPcon/LinuxCon

Arrange travel for LDAPcon/linuxcon in Portland, OR next week.

September 10, 2009

Encrypted $HOME Now Offerred at Installation

I'm pleased to announce that the Ubuntu Karmic Alpha5 image now offers home directory encryption as an option to all installing users!



We introduced Encrypted Private Directories in the Ubuntu 8.10 release, using eCryptfs (an enterprise cryptographic filesystem in the Linux kernel) on $HOME/Private. This release helped "prove" eCryptfs, and helped us identify and fix a number of issues. This new approach to encrypted private data in Ubuntu provided a safe folder where users could store confidential information, automatically mounted at login, and unmounted at logout.

In Ubuntu 9.04, we retained the Encrypted Private Directory feature, but additionally offered Encrypted Home Directories to advanced users, through the alternate installer and a special boot parameter. This release generated quite a bit of interest in the feature and a healthy user community. Many, many thanks to the Ubuntu users and developers who used this feature, helping to file and fix bugs along the way.

So far in Ubuntu 9.10, we have:

  • fixed a number of bugs and usability issues (changelog)

  • provided AppArmor rules

  • enabled the shell scripts for localization/translations

  • and most importantly, set up encrypted swap in the installer if you enable home directory encryption


I believe Ubuntu now provides the most user-friendly personal data encryption solution in the industry.

So secure your data in Ubuntu! Get those Karmic home directories encrypted!

:-Dustin

p.s. I authored an article for Linux Magazine that should be published in an upcoming issue discussing the technology in much greater detail. Stay tuned!

September 09, 2009

qemu-kvm-0.11~rc2 uploaded to Karmic

The upstream qemu and kvm projects have released the second release candidate of qemu-kvm-0.11, the stable series of the accelerated virtualization hypervisor for Linux.

I have merged and uploaded this package for Ubuntu Karmic. Please test on your systems, and file bugs with: ubuntu-bug qemu-kvm

Karmic is rapidly approaching Beta, RC, and GA status. Please help test kvm!

Cheers,
:-Dustin

Server Team 20090908 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

UEC images

There was a discussion on how to track bugs affecting UEC images. The outcome was to use the “uec-images” tag. ubuntu-bug and apport should also use it when filling bugs from an UEC image.

Test cases have been updated to include an EC2 test of the UEC images. That should be enough for now to make sure the UEC images are working correctly.

ACTION: smoser to tag existing UEC image bugs with “uec-images”

ACTION: mdz to follow up on ubuntu-bug/apport for uec images

EC2 AMIs

The release process for EC2 images was reviewed in view of Alpha6 scheduled for next week.

ACTION: soren to ensure that smoser can update the UEC publishing scripts

ACTION: smoser to add MD5SUMs for UEC images

ACTION: soren to add manifest files for UEC images

ACTION: smoser to open dialog with IS about automated publishing to EC2 and agree on a plan

ACTION: soren to automate updating of ec2-version-query

ACTION: soren to publish ec2-version-query in a more appropriate place

The state of the EC2 kernel was also discussed. zul reported that a 2.6.31-rc6 kernel based on the Xen patches had been successfully booted on ec2. Merging the patch into the karmic tree was under way.

ACTION: mdz to confirm ec2 kernel status for alpha 6

ACTION: smoser to add ec2-images tag to the relevant bugs

ACTION: mdz to see that bug documentation is updated for uec-images, ec2-images tags

ACTION: nijaba to fold #ubuntu-ec2 and #ubuntu-cloud into #ubuntu-server

Packaging and integration of Eucalyptus 1.6

As agreed with the Eucalyptus team relevant bugs should be tagged with ‘eucalyptus’. soren announced that Eucalyptus 1.6 had landed in Karmic. The installer experience has been reviewed by the Eucalyptus team and bugs have been filed. cjwatson is looking at them.

ACTION: soren to triage all eucalyptus bugs, and use the ‘eucalyptus’ tag for bugs which should be escalated to the eucalyptus team

Alfresco appliance

The main blocker to produce an appliance was the dependency on Sun jdk. Its EULA needs to be accepted by the end user when deploying the appliance.

ACTION: kirkland to build a proof of concept alfresco appliance

Appliance store

niemeyer reported that most of the code had been written. The store UI has already been integrated in Eucalyptus 1.6.

ACTION: mathiaz to get niemeyer’s proxy code packaged

Canonical application support

ACTION: zul to ensure rabbitmq-server gets reviewed and promoted

Directory items

2.4.18 has been released on Sunday and the FF Exception has been granted.

ACTION: mathiaz to upload openldap 2.4.18

CIM/WBEM

mathiaz reported the spec was implemented. More testing is welcome.

KVM-QEMU testing

kirkland noted that most of the bugs on kvm / qemu were reported after RC. He asked for ways to increase the testing of kvm in the development release.

ACTION: kirkland to speak with marjo about how to get qemu-kvm tested prior to release (and more generally server applications like it)

Server developer team in LP for ArchiveReorganisation

Package sets are available in Launchpad, and cjwatson was looking for the appropriate teams to have privileges on them. But there didn’t seem to be an appropriate team for Server Edition. If the server team wants to take advantage of the ability to provide upload privileges for server bits without requiring core-dev, somebody will need to sort it out.

ACTION: mathiaz to get a server dev team set up in LP and work with cjwatson to get it set up for archive reorg

Assigned and to-be-assigned bugs

The list will be used to keep track of active bugs in the team. The list needs to be reviewed and kept up-to-date.

Progress on Roadmap

The ServerTeam Roadmap is not up-to-date.

ACTION: ttx to update server team Roadmap to reflect current projects

Asterisk

The Debian VOIP hasn’t responded yet.

ACTION: Daviey to call for testing of Asterisk 1.6

SRU review

ACTION: mathiaz to produce a list of accepted bugs for packages related to the ubuntu-server team.

Agree on next meeting date and time

Next meeting will be on Tuesday, September 15th at 15:00 UTC in #ubuntu-meeting.

September 01, 2009

Server Team 20090901 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

FFE for bacula

ivoks reported that bacula 3.0.X was available from the ubuntu-bacula team PPA.

Asterisk

Daviey announced that the Asterisk stack in karmic had been updated to 1.6. Discussion is still going with the Debian maintainers about the dkms version of dhadi. Testing is welcome.

jmdault added that asterisk-addons needed a Feature Freeze Exception. A bunch of other packages (eg asterisk gui) will be made available from the ubuntu-voip team PPA.

ACTION: Daviey to follow up with the Asterisk Debian maintainers about adopting dkms for dahdi

ACTION: Daviey to call for testing of Asterisk 1.6

SRU Weekly review

The different bug lists were reviewed during the meeting.

ACTION: mathiaz to produce a list of accepted bugs for packages related to the ubuntu-server team.

MIR for ubuntu-fortunes and default install

nijaba proposed to file a MIR and get the fortunes-ubuntu-server package installed by default.

ACTION: nijaba to write up a MIR and file a FFe for fortunes-ubuntu-server

Agree on next meeting date and time

Next meeting will be on Tuesday, September 8th at 15:00 UTC in #ubuntu-meeting.

August 27, 2009

Server Team 20090825 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Asterisk

jmdault reported that pwlib and openh323 had the right version in karmic since Friday. The rest of the Asterisk packages have been uploaded to his PPA and are waiting for sponsorship. The Asterisk wiki specification has also been updated to reflect the state of the stack in Karmic.

Update Ebox to 1.3

zul reported that ebox had been updated to 1.3 in Karmic universe.

SRU weekly review

The lists of nominated bugs for each supported release as well as the list of last-week-fix-released bugs have been reviewed during the meeting. Relevant bugs that were SRU worthy have been accepted.

ACTION: mathiaz to produce a list of accepted bugs for packages related to the ubuntu-server team.

FFE for bacula

There was some discussion on whether time should be spend on getting Bacula 3.0.X merged from unstable into karmic. It was decided that it was too late for karmic and thus deferred to the next release cycle. Providing the new version in a karmic PPA would also be useful.

ACTION: bacula 3.0.X to be made available from a PPA from the bacula team.

UEC/EC2 images

mdz asked about the status of EC2 images. Karmic images are now built on a daily basis and were included in Alpha 4.

  • bug 376740 tracks the failures of the kernel on Karmic AMIs.

  • bug 398568 is tracking Jaunty AMIs failing to boot.

The Ubuntu package archive servers inside EC2 are alive. Both Jaunty and Karmic AMIs are using these by default.

Agree on next meeting date and time

Next meeting will be on Tuesday, September 1st at 15:00 UTC in #ubuntu-meeting.

August 24, 2009

A Statistical Analysis of Potential PowerNap Energy Savings


I was asked recently how much energy savings administrators might actually expect from an Ubuntu Enterprise Cloud, powered by Eucalyptus and using PowerNap.

I have blogged before about how much I enjoy mathematics and statistical analysis, and this is in fact another exciting, hard question!

The question was presented as the following hypothetical...
Given 10 quad-core servers in a cloud running 40 virtual machines. 20 virtual machines are simultaneously discarded. What power savings would you expect from PowerNap?
This is actually a rather complex combinatorics problem. I'll restate it as such:
Given 10 buckets, each of which can hold 0 - 4 items, how many ways can 20 items be distributed among those 10 buckets? Furthermore, what is the statistical distribution of empty buckets?
It's been 10 years since my university statistics classes, and my brain melted trying to derive the permutation formula. But a big thanks to my good friend Matt Dobson for solving that! If you're interested, you can view his combinatorics formula here.

It took me about an hour to hack a Python script that could empirically solve this problem by brute force, generating a comprehensive list of all of the permutations. You can now download and run /usr/bin/powernap_calculator provided by the powernap package in Karmic.

Here are the results running on the parameters above:

$ time powernap_calculator --hosts 10 --guests-per-host 4 --guests 20
Calculating...99%

In a cloud with [10] hosts, which can handle [4] guests-per-host, currently running [20] guests,
you may expect the following:

[ 5.2%] likely that [0/10] of your hosts would powernap, for a [0%] power savings
[ 27.5%] likely that [1/10] of your hosts would powernap, for a [10%] power savings
[ 42.5%] likely that [2/10] of your hosts would powernap, for a [20%] power savings
[ 21.8%] likely that [3/10] of your hosts would powernap, for a [30%] power savings
[ 2.9%] likely that [4/10] of your hosts would powernap, for a [40%] power savings
[ <1%] likely that [5/10] of your hosts would powernap, for a [50%] power savings

The overall expected value is [19.0%] power savings.

real 0m46.919s
user 0m46.227s
sys 0m0.276s


So at this snapshot in time, when your cloud suddenly dropped to 50% utilization, the expected value is a 19% power savings. Note that expected value is a very specific statistics term. Wikipedia says:
In probability theory and statistics, the expected value (or expectation value, or mathematical expectation, or mean, or first moment) of a random variable is the integral of the random variable with respect to its probability measure. For discrete random variables this is equivalent to the probability-weighted sum of the possible values, and for continuous random variables with a density function it is the probability density-weighted integral of the possible values.
Right, so we might expect about a 19% power savings for this random moment in time, when we simultaneously reduce our utilization by 50%.

However, if we observe the cloud over time, and with perhaps more realistic usage patterns, the distribution should be much better than random. VMs will come and go at more staggered intervals than simultaneous destruction of half the instances.

And you will have another tremendous factor working in your favor -- Eucalyptus' greedy scheduling algorithm. This ensures that each time a new VM is launched, it will land on a system that's already running. This is known as an annealing system -- one that's constantly correcting itself -- since under-utilized systems will automatically powernap, and new VMs will fill in the gaps on running hardware. Pretty cool, I think.

So I'm curious...
  • What does (or would) your cloud look like?
  • How many -h|--hosts, -p|--guests-per-host, and -g|--guests do you usually have?
  • What does powernap_calculator say about your parameters?
  • Post your results!
Note that the powernap_calculator algorithm is exponential, O((p+1)h), so large values of (p, h) will take a very long time to compute! I'm totally open to code review of powernap_calculator and any performance enhancements ;-)

In my cloud, I have 8 dual-core hosts. I typically limit my usage to 2 guests-per-host. And I rarely run more than 4 VMs at a time.
$ powernap_calculator -h 8 -p 2 -g 4
In a cloud with [8] hosts, which can handle [2] guests-per-host, currently running [4] guests,
you may expect the following power savings:
[ 26.3%] likely that [4/8] of your hosts would powernap, for a [50%] power savings
[ 63.2%] likely that [5/8] of your hosts would powernap, for a [62%] power savings
[ 10.5%] likely that [6/8] of your hosts would powernap, for a [75%] power savings
The overall expected value is [60.5%] power savings.
My servers run at about 80 Watts fully loaded. My electricity is about $0.10 per kilowatt-hour. So a year's worth of electricity with all 8 servers running all of the time is 8 * .08 KW * 24 hr/day * 365 days/year * $0.10/KW-hr = $560/year.

I like the prospects of saving approximately 60.5% of that, or $339/year!

So how does the calculator work?

Let's use a slightly smaller example: hosts,h=4, guests-per-host,p=2, guests,g=3

Since each system can hold 0, 1, or 2 virtual machines, we're going to use a base-3 number (which is p+1). And we're going to generate all possible 4-digit base-3 numbers (which is (p+1)h). In our case, that's 34 or 81 scenarios to consider. For each of those scenarios, we convert the decimal integer to a list of each of the digits 0-2, and sum the list, throwing out any "invalid scenarios", where the sum does not add up to the target number of guests, 3, g. Our valid scenarios is actually much smaller, the following 16:
[2, 1, 0, 0]
[1, 2, 0, 0]
[2, 0, 1, 0]
[1, 1, 1, 0]
[0, 2, 1, 0]
[1, 0, 2, 0]
[0, 1, 2, 0]
[2, 0, 0, 1]
[1, 1, 0, 1]
[0, 2, 0, 1]
[1, 0, 1, 1]
[0, 1, 1, 1]
[0, 0, 2, 1]
[1, 0, 0, 2]
[0, 1, 0, 2]
[0, 0, 1, 2]
Now, of these, we are interested in how many 0's are in each row, as this indicates a host that has no guests, and can therefore be powernapped.

Our calculations yield: [0, 4, 12, 0, 0], or:
  • 0 possible scenarios have no unused hosts
  • 4 possible scenarios have 1 unused hosts
  • 12 possible scenarios have 2 unused hosts
  • 0 possible scenarios have 3 unused hosts
  • 0 possible scenarios have 4 unused hosts
From this we can generate the following probabilities:
[ 25.0%] likely that [1/4] of your hosts would powernap, for a [25%] power savings
[ 75.0%] likely that [2/4] of your hosts would powernap, for a [50%] power savings
And weighting these probabilities, we can generate the expected value:
25%*.25 + 50%*.75 = 43.75%
The overall expected value is [43.8%] power savings.

Comments appreciated!

:-Dustin

PowerNap in the Podcasts




Howdy all-

If you don't already follow the Ubuntu Podcast and the Ubuntu UK Podcast, you really should! Both are community driven multimedia presentations that cover the news in the Linux and Ubuntu communities. These guys provide a tremendous service to the world of Ubuntu.

Nick Ali and Josh Chase discuss PowerNap in the Ubuntu Podcast Episode #31. They did a good job explaining the design, uses, and advantages of it. Give it a listen...

And the Ubuntu UK Podcast guys (Ciemon Dunville, Alan Pope, Tony Whitmore, Dave Walker) interviewed your humbled author in S02E09, The Dimensions of Time. We talked about the Ubuntu Server, PowerNap, KVM, and eCryptfs.

Enjoy!

:-Dustin

Introducing PowerNap!

Ladies and gentlemen, I'm pleased to announce an exciting, new open source project from Canonical for the Ubuntu Server -- PowerNap!

Mark's Karmic Koala announcement alluded to this work when he wrote:
A savvy Koala knows that the best way to conserve energy is to go
to sleep, and these days even servers can suspend and resume, so imagine
if we could make it possible to build a cloud computing facility that
drops its energy use virtually to zero by napping in the midday heat,
and waking up when there's work to be done. No need to drink at the
energy fountain when there's nothing going on. If we get all of this
right, our Koala will help take the edge off the bear market.
I have just uploaded PowerNap to Karmic, and we are well on our way to integrating the technology into the 9.10's Ubuntu Enterprise Cloud.

Actually, I spent last week in sunny Santa Barbara, California working with Dan Nurmi, of Eucalyptus Systems. We shot some amateur digital videos of Ubuntu Karmic Servers, PowerNap/PowerWake, Eucalyptus, and a Watt meter in action. I'll get those posted soon!

I'll go into much deeper technical detail on the design and implementation of PowerNap over the next few weeks in subsequent posts, but I'll give an overview here...

How Does It Work?

PowerNap operates sort of like a screen saver for servers. Besides watching the console and terminals for keyboard activity, it also watches the system's process table for activity.

An administrator defines a list of regular expressions describing some critical MONITORED_PROCESSES that should be running. When powernapd notices that all of the MONITORED_PROCESSES have been absent from the process table for some configurable ABSENT_SECONDS, powernapd emits a warning to all users of the system that it will run powernap, unless canceled within the next GRACE_SECONDS.

Sample Configuration

In the Ubuntu Enterprise Cloud case, the configuration file, /etc/powernap/config, might look something like:
MONITORED_PROCESSES = [ "^/usr/bin/kvm " ]
ABSENT_SECONDS = 300
GRACE_SECONDS = 60
Thus, if no instance of kvm runs for 5 minutes, then the system will emit a warning, and powernap after a 1 minute grace period.

PowerNap Now!

Alternatively, a system administrator can force the system to powernap immediately by either running /usr/sbin/powernap directly, or sending powernapd the "now" signal with: service powernap now. In fact, this is what the Eucalyptus Node Controller does, such that it can maintain state, and directly control its managed nodes.

What constitutes a powernap?

So, powernap will first check if /etc/powernap/action is executable, and if so, it will run that file. This will allow you, as an administrator, to run any arbitrary script or program of your design when powernapd determines that your server has become idle. Your script could send an email, for example.

echo "Inefficient server, wasting energy" | mail Al_Gore@example.com
But in the default case, powernap will check if your server supports suspend-to-ram, and if so, it will pm-suspend your system. Otherwise, it will suspend-to-disk, or power the system off, depending on the sleep states supported by your hardware.

Slick, huh? :-D

Beyond the Cloud

While PowerNap is bespoke for the Ubuntu Enterprise Cloud, I have implemented it in manner that I hope is generically useful.

In fact, I'm currently using PowerNap on my Mythbuntu front ends! My configuration currently looks something like:
MONITORED_PROCESSES = [ "mplayer", "vlc", "xine", "mythfrontend.real", "xmms", "gthumb" ]
ABSENT_SECONDS = 240
GRACE_SECONDS = 60
My hardware supports S3 suspend-to-ram, so this is great! If 4 minutes go by, where I'm not running any of my media players (mplayer, vlc, xine, mythfrontend.real, xmms), I'm given a 1 minute grace period, and then my system suspends. I have configured wake-on-usb and wake-on-lan in the BIOS, so I can resume the system in a couple of seconds either by tapping a key or sending a WoL magic packet.

But in the mean time, I've reduced the power consumption of 4 systems by 90%, for most of every day while I'm not directly using MythTV!

What about Waking Systems?

Which brings me to PowerNap's kid brother...PowerWake. /usr/bin/powerwake is another Python script. This script is designed to be a smarter, remote waking utility. Currently, it supports wake-on-lan, but it will eventually support other mechanisms, such as IPMI, and perhaps NUT.

With respect to wake-on-lan, it's "smarter" than some other wake-on-lan utilities because it uses a hierarchy of cache files, configuration files, and the current arp table, such that you can wake a system by MAC address, or IP address, or hostname. I find this far more convenient than trying to remember or look up MAC addresses. powerwake respects static configuration in /etc/ethers and maintains a dynamically learning cache of known MAC addresses in /var/cache/powerwake/ethers.

Interesting?

I'm eager to hear what other uses you might have for PowerNap and/or PowerWake for your data centers, basements, and living rooms!

Saving a few Watts,
:-Dustin

August 22, 2009

qemu-kvm: call for testing

If you're running Karmic and you use KVM, I'd appreciate your help testing the qemu-kvm package.

sudo apt-get install qemu-kvm

And then just use KVM as you normally do. Please file bugs at:
Thanks,
:-Dustin

August 18, 2009

Server Team 20090818 meeting minutes


Here are the minutes of the meeting. They can also be found online
with the irc logs here.

Asterisk

jmdault reported he had built all packages relevant to the Asterisk stack in his PPA. This update would bring Asterisk to the latest version: 1.6. Testing is welcome.

ACTION: jmdault to ask sponsorship for pwlib -> openh323 -> dahdi* -> asterisk* packages

Update Ebox to 1.3

foolano stated that packages were already in a PPA and debdiffs had been attached to bugs.

ACTION: zul to review ebox 1.3 packages

SRU weekly review

mathiaz added the list of nominated bugs relevant to packages interesting to the ubuntu-server team to the list of bugs fixed during last week. Both lists were reviewed during the meeting and relevant bugs were accepted for SRU. mathiaz reminded that the updated SRU process was documented in the Ubuntu Server Knowledgebase.

ACTION: ttx to go throught the list of nominated bugs and decline them

libmysqlclient-dev transition

mathiaz announced that MySQL 5.1 had been promoted to main on Monday and that transitioning build dependencies (libmysqlclient15-dev to libmysqlclient-dev) had started for packages in main. MySQL 5.0 will stay in universe for karmic and will be removed from the repository during the next release cycle.

Server tips : implementation

Daviey reported he had committed 32 new tips to the ubuntu-server-tips repository. Translations also started. nijaba offered to review submitted tips. Work for integrating the package with update-motd is also underway with some help from kirkland.

Agree on next meeting date and time

mathiaz reminded everyone that Feature Freeze is scheduled for next week.

Next meeting will be on Tuesday, August 25th at 15:00 UTC in #ubuntu-meeting.

Atlanta LinuxFest

Wanted to drop a quick note about ALF...

I'll be speaking at the Atlanta LinuxFest, talking about the Ubuntu Kernel. This will talk about the team, how we develop & maintain the Ubuntu Kernel.

In fact there are several Canonical folks talking at ALF.
  • Steve Conklin - Kernel Debugging
  • John Pugh - The Weather Ahead - Clouds
  • Ken VanDine - Ubuntu Desktop Experience
  • Rick Clark - Ubuntu Server
Along with all the speakers noted above, members of the Ubuntu Kernel Team will be conducting a driver writing session. This is based on GregKH's original driver writing presentation. We will be using a USB thermometer as the hardware and the object is to write a working driver that will get the current temperature.

Along with ALF there will be a Ubucon (Ubuntu User Conference). The Ubuntu Kernel Team will be holding a "Is your hardware ready for Karmic" workshop. We will have USB sticks loaded with a testsuite that will test most of the new Karmic hardware features like Kernel Mode Setting (KMS), grub2, net/notebook hotkeys, web cams, audio and the like. This is non destructive to the hard disk and will let folks know in advance about what the Karmic experience will look like on their hardware. If items fail to work we will file bugs on the spot with the proper debugging info attached to the bug.

I want to give a huge shout out to Manjo Iyer & Ronald Fader who have spent lots of time and hard work to make the testing happen and to John Johansen for putting together the driver writing session! Great work guys

More on ALF as it gets closer.

~pete

Split Routing Over 2 DSL Lines

Its been a bit since I've blogged last. Most of it is due to moving. We moved from Raleigh NC to a small town in the western part of NC called Union Mills. The good thing is we are on a farm, we have lots of land, space, fresh air... however the Net connection just plain sucks.

I'm 2.3 miles from the CO, so the best I can get with DSL is a 3m/384k connection thru AT&T. So I called AT&T commercial and was assured I could be 2x bonded ADSL lines that would give me effectively a 6mb connection. Guess what? The fibbed. In the end I had to settle for 2x ADSL lines not bonded, and I elected for the non-commercial option since it was far cheaper.

Now came the big question how to best effectively use em'. After hitting up Google I found lots of interesting solutions. Most were very complex, routing various protocols out this interface or that.... I wanted something that would give me the closest to a load balanced connection as possible.

I'm using a old HP Desktop with 3 network cards as my gateway router running Jaunty 9.04 Server Edition.

Below is what I came up with using ip route and iptables. I put in bogus IPs to keep it as real as possible.

First I added two routing tables to the /etc/iproute2/rt_tables file:

1 line1
2 line2

Then I found a script on the net and used it as a template and hacked it up as so:

#!/bin/bash

# DSL Lines are IF0 & IF1, IF2 is local net.
IF0=eth0
IF1=eth1
IF2=eth2

# IP Addr on Gateway matching interfaces above
IP0=192.168.1.1
IP1=192.168.2.1
IP2=172.31.0.1

# DSL IPs
P0=192.168.1.254
P1=192.168.2.254

# Network addresses
P0_NET=192.168.1.0
P1_NET=192.168.2.0

# Routing table entries
T1=1
T2=2

# Set up routes
ip route add $P0_NET dev $IF0 src $IP0 table $T1
ip route add default via $P0 table $T1
ip route add $P1_NET dev $IF1 src $IP1 table $T2
ip route add default via $P1 table $T2

ip route add $P0_NET dev $IF0 src $IP0
ip route add $P1_NET dev $IF1 src $IP1

# Set up default route to balance between both interfaces
ip route add default scope global nexthop via $P0 dev $IF0 weight 1 \
nexthop via $P1 dev $IF1 weight 1

# Add the rules for the routing tables
ip rule add from $IP0 table $T1
ip rule add from $IP1 table $T2

# Now for the masq bits
iptables -F
iptables -X
iptables -t nat -F
iptables -t nat -X
iptables -t filter -F
iptables -t filter -X
#iptables -t nat -A POSTROUTING -o $IF0 -j MASQUERADE
#iptables -t nat -A POSTROUTING -o $IF1 -j MASQUERADE
iptables -t nat -I POSTROUTING -s 172.31.0.0/24 -j MASQUERADE

# Turn on ip_forwarding
echo 1 > /proc/sys/net/ipv4/ip_forward

iptables -A FORWARD -i $IF2 -s 172.31.0.0/24 -j ACCEPT
iptables -A FORWARD -m state --state ESTABLISHED,RELATED -j ACCEPT

# Flush routing every four hours
echo 144000 > /proc/sys/net/ipv4/route/secret_interval

It works like this... Each time a connection is established outbound, the server will decide which interface to route it out of. This should be a rough 50/50 split. The server decides based on congestion and a few other factors. Only one connection can only ever be established over the same route.

If I was start a download, it will only ever utilize a single connection, another download from the same IP will also utilize the same connection. This is due to the kernel having already cached the route. If a 3rd download to a new server was to be started it will likely be established over the second connection due to the first route being congested and the second route is idle.

The route will be flushed 4 hours after the first download completes, then a new decision with be made on how to contact the original server.

I've been using this for a bit now and it seems to do what I need it to do... give me a faster response time when I have intensive net operations going on, like vonage & rsync's to offsite machines.

By no means am I a iptables or routing guru. If anyone has any other suggestions or better ways to do it I'd love to hear them.

~pete

Ubuntu Server Planet

Ubuntu Server Planet is a window into the world, work, and lives of Ubuntu Server Team contributors all over the world

http://www.ubuntu.com/server

Ubuntu World Wide

Ubuntu Worldwide

Subscribe


Site Last Updated:
November 30, 2009 06:14 PM (UTC)