The Innerworkings of a Geek

Environmentalism, Politics, and 4wheeling

2009-06-16T18:08:00.001-07:00

Many of you know I'm an avid 4-wheeler. Its a fun and challenging hobby, and it includes a great community of people.

For some reason, there is growing pressure from environmentalists, hikers, and MTN bikers to ban ORV use, eliminate or downsize ORV parks, and forbid the creation of new parks. Many believe ORV use is a scourge on our federal and DNR land and must be eliminated.

Well I'm going to point out two flaws in your logic:
1) Its not that much worse than hiking and biking trails, especially in comparison to logging, development, and wildfires
2) By taking away ORV parks, you've just invited wheelers into your own backyard. Instead of wheeling on sanctioned trails, they'll be wheeling on YOUR trails.

Nevertheless, there are groups that won't be happy until 4x4 is illegal. They claim trash, mud and damaged trees as reasons. We respond by having bi-annual cleanup parties where hundreds of people show up to clean up the trails. We also advocate through 4x4 clubs responsible wheeling, going on legitimate trails etc.

But all this effort of closing Reiter is only going to make things worse

Where are the wheelers going to go if Reiter is closed?
Are they going to just sell their $8,000-90,000 rigs and find a new hobby? No.
Are they going to drive hundreds of miles to wheel every weekend? No.
Are they going to find trails off secluded DNR/Forest service road? DING DING DING!

So now, you environmentalists and politicians have moved a contained area, wheeling within a 40,000 acre Park, to a bunch of random 'Fight club-esque' places that no one reveals so they don't get shut down.

Don't believe me? North Fork was the only ORV park in Whatcom county. It was shut down in 2006. Do you think everyone went to canada or walker valley? HA. I'll just say that there is a large contingent of trails wheeled every weekend in Whatcom county, ranging from near Bellingham all the way out to Mt. Baker.

Are work parties to cleanup trash created in these areas? NOPE
Are there work parties to create culverts and safe water passages? NOPE
Is there any coordination between clubs to make a manageable trail system that leaves a small footprint on sensitive areas? NOPE

And there you go, you've just moved a contained group that exists in one area to a fragmented mess spread around western Washington. And the ONLY way to totally eliminate this is by gating off everything. If that happens, then say goodbye to hiking and biking trails, as the trail-heads will be miles from the gate.

So this is my message to any environmentalist reading this: You WANT to support ORV parks.
When you support ORV parks....
....you're supporting a managed system of trails that are being maintained and patrolled by fellow ORV users and law enforcement.
....you're supporting people who have vested interest in keeping that land clean and enjoyable by all.
....you eliminate or drastically reduce the problem of illegal wheeling in the forest where YOU might be hiking/biking/climbing.

This is my message to politicians who make the policies: Supporting complete ORV parks will help stop illegal dumping, and illegal trail systems
1) Rate trails. To me, it seems like DNR would be more legally sound if they rated trails. If some dumbass gets his rig stuck, flops, etc and there is ample signage to say its difficult, then I'd think they'd be covered in a lawsuit.
2) With rated trails, don't make all of them watered down, if the trails aren't hard for those with 90k buggies, they will go elsewhere.
3) Use the vast 4wheeling community to build, manage, and maintain trails. This is working successfully at walker valley.
4) Create a framework for funding so the 4wd community can allocate funds from their own fundraisers into grants to build trails in ORV parks.
4) get NOVA funds back into the proper places

I find ORV parks to be very similar to Ski resorts, in both environmental impact and trail design. So why can't we get just a few good ORV parks in this area?

Earth Hour: Turn your brain off for 60 mins!

2009-03-28T21:51:00.000-07:00

I was going to write a Twitter or Facebook quip about earth hour, but I just couldn't fit everything in. So here comes a blog post.

First, not enough people are thinking critically about this. Most people acknowledge that Earth Hour does nothing to conserve energy. Even Edward Norton, the 'US Ambassador' for EA acknowledged this on Larry King tonight.

Then, I'm watching more of Larry King and there is a PSA for Earth Hour. There we have Alanis Morissette ON A PLANE advertising what you need to do for earth hour. wow.

And lastly we have the guy who leads the WWF, who is pushing all these environmental actions, with little scientific research to back it up. None of these three people are scientists, and none of them acknowledge the debate regarding climate change. There is ALWAYS a debate, and making assumptions like they do is similar to 'soft fascism' (This article is definitely libertarian/free market biased, however it makes some good points.)

And all this boils down to Penn and Teller's episode on 'Environmental Hysteria' ... And if you watch it, you'll notice all these 'environmental followers' sign a petition to ban water. Thats right, water.

Climate change is real, no doubt about that. But the way environmentalists are approaching its solution are based on politics, not science. Many advocate turning back the clock on the industrial revolution, or charging for carbon credits, which ultimately harms the poor and middle class. Technology and Science is going to be what gets us out of this. Better solar and wind energy plants, better battery technology, and better transportation technology. The government needs to invest and support companies and universities, not taxing John doe and Corp X. There is a HUGE market for electric cars, and sustainable power, when the technology is good enough to compete. This will happen, I think sooner than later. Until then, if you want to live in an unpowered cave, or drive veggie cars, or spend a whole bunch of money on solar power and make your house energy efficient, kudos. Hybrids don't count ;-)

Think about it this way: Babies are bad for the environment. And if you look at the Earth Hour movement stats, it'd take just one person to not have an extra kid (living in an industrialized nation) to do the same as Earth Hour. Perhaps people should reduce having kids to one or two. That would be more environmentally friendly than almost any other suggestion offered out there.

Earth Hour is about being smug at best, and trying to push dubious science as law at worse. So smarten up and don't follow blindly. Turn your lights on for Earth Hour!

This youtube video pretty much sums up my view: http://www.youtube.com/watch?v=Ifci-W2Cs1o

Development environment notes

2009-03-04T08:33:00.000-08:00

Here is a good raw list of things to have when developing drupal sites....

Sacha from IBM -- http://bit.ly/drupal25

Spend the time to make a tool, if it make sense
Read editor documentation
Some browser plugins Tamper Data, iMacros, Drupal for Firebug (plugin and drupal module), trace module, drubuntu (for multisite is very useful)

Simpletest! I've got to research simpletest more

source code management - using svn, git, etc
--check out the whole source tree
--check in clean source for third party
--organize the sites/modules dir for development, etc

use update functions in the custom module install code.
create svn branches for production and dev
use make or ant to build your system
* sql --- connect db
* backup --backup db
* restore -- restore db
* clearcache -- clear cache
* tags -- rebuild VI tags
* doc -- update doxygen docs
* test -- run project tests
* parse settings.php file to get variables

Use the drupal shell! http://www.drupal.org/project/drush
vimperitor -- allows vim shortcuts for firefox
Use Virtual document root - apache can be the SVN checkout
http://httpd.apache.org/docs/2.2/mod/mod_vhost_alias.html#virtualdocumentroot
unfuddle
Write install profiles for projects
Aegir

LTSP vs Hybrid (Diskless node) clients

2009-02-27T11:49:00.000-08:00

This week, we rolled out our new deployment of linux workstations. These systems are running SLED11 RC4, with Firefox, OpenOffice, etc. Featurewise they are the same as our OpenSUSE 11 LTSP machines. However, they're now Hybrid orDiskless nodes

Some history....
When I got into deployment of LTSP machines, I thought thin-computing would be pretty straightforward. At the beginning it was. We had only a few users, they were not power users (only Firefox, OpenOffice) However, the more users, the more people wanted to use them as normal workstations. People in the LTSP channels boasted how easy they were to administer, the reliability of just servers to maintain, etc, etc. I was excited when I got my first crack at a deployment. But would it match up to a normal workstation? In short the answer was no.

Why did we choose LTSP?

1) Reduce Costs
By having two servers operating for 40 users, we should be able to have 40 thin clients priced below $300/CPU. When I approached dell for a thin client terminal, they wanted $625 for just the CPU! I said no thanks, and instead went the custom route. We spent between $275-$225/ea CPU, plus ~$200 for Samsung 19" LCDs. Prices have come down on LCDs since then.
2) Reduce management
Two servers are much easier to update than 40 clients.
3) Workstation security
Data is always stored on redundant, backed up servers; not the client. The client also doesn't have local root and cannot install local apps, so the machine won't be infected with viruses.
4) Reliability
If the local machine crashed, it can easily be swapped out for another machine. Since the server is the only system being maintained, it should be more reliable.

What went wrong with thin-client computing for us? I wrote about it more here: http://www.fcdnet.org/japerry/2008/12/good-bye-ltsp.html

Why it failed:
1) They're slow.. painfully slow. You'd think with 2 8-core, 8GB, Raid 0 servers you'd have fast clients. When compiling, yah they're fast. Otherwise, they are really slow. Monika mentions that her work also uses thin-clients, only the windows variant. They too are slow and have reliability problems.
2) Unreliable. If the server has issues, all users are affected. We did everything possible to keep them running, but we hit a major unexpected roadblock here: Many apps are just not tested or designed to run in a multi-user environment. Especially file/network protocols. Novfs, NFS, NSS, SMB all one issue or another with multiple users on one machine. And with other apps we find 'odd' issues that just don't occur on single machines.

So what about a diskless node solution? Keep the drive remote and treat the client similar to a liveCD.....
KIWI-LTSP project is much more than LTSP. It also is for image automation. although still in beta, its used in many production processes. SuSE releases its Live/Install CD/DVDs using the kiwi imaging system. Suse just announced SuSE studio, which is a web based service to create custom disk and vmware images, essentially a front-end to the kiwi imaging system. But there are other things you can do with kiwi as well, including boot strapping a NFS, NBD, or AoE squashfs system.

Our Solution: AoE diskless nodes, using KIWI.
Cyberorg recommended I take a look at AoE support in KIWI. Work was done recently to allow AoE exports to mount in kiwi. Ata over Ethernet. The nice thing about AoE is that its performance is almost that of iSCSI. The disadvantage (some would say) is that you can only export it on the local lan, since its not routable. for our situation, this is fine, and good because it provides an extra layer of security. Our hybrid solution sits behind its own VLAN.

The other major advantage of AoE and KIWI is ease of image updates. Once the kiwi netboot initrd is setup, you can change your images without needing to rebuild kiwi. For our deployment, I have a dedicated hard drive for images. This allows us to fully test (IE reboot into) a snapshot of our image. Once testing is done, I reboot into my deployment machine, run a few custom scripts to sanitize /tmp, /var/log, /etc (get rid of udev, fstab, etc), and export the whole drive as a squashfs.

 mksquashfs /mnt/staging /srv/kiwi-images/`cat /mnt/staging/etc/suse-release`-`cat /mnt/staging/etc/kiwi-label`-`cat /srv/kiwi-images/version-incr`.img

Once this is done, i can deploy the new image.

losetup -a

losetup -d /dev/loopN (where N is the loop devices already mounted

losetup -v -f /srv/kiwi-images/`cat /mnt/staging/etc/suse-release`-`cat /mnt/staging/etc/kiwi-label`-`cat /srv/kiwi-images/version-incr`.img

losetup -a // verify what loop device the image is deployed on
vbladed 0 1 eth1 /dev/loopN

One drawback to the process above is the need for everyone to shutdown their systems before the image can be deployed. There is a workaround to this. If you don't delete the previous image, you can change the pxeboot.cfg file to indicate the next loop device (probably /dev/loop1). Then, whenever someone reboots, they'll get the newest image, but currently logged in users aren't affected. Depending on your user base, you may want to incorporate a cron job to forcibly shutdown/reboot the machine (for example saturday morning, 3am). Since these are linux machines, some of our LTSP clients have had an uptime near 60 days. If you decide to deploy an image, and then delete the old one a month later, you may find two or three users calling the helpdesk wondering why they get 'cannot execute file: input/output error' and cannot reboot their machine without hard shutting it off.

Another solution is to create one base image, and perform updates entirely from scripts. For servers I could see this being useful, but for clients its a little more sketchy. Right now the image deployment allows a user to be at the login screen exactly one minute after they press the power button (and 20 seconds of that is before pxeboot even starts!) If we ran zypper to install custom packages, the boot time would increase considerably. So we install all needed packages on the system, and use configuration scripts to decide if something should start on boot.

Custom configurations
In LTSP, we could deploy the lts.conf file, which would setup sound, printers, and other special devices an individual user may have. Since we're no longer using LTSP, I needed to come up with a different solution. I created a script that gets called on boot that copies or appends files based on the hostname. This allowed me to get custom printer.conf files for users, which would append to the company-wide printers already installed on the image.

Home directories, stateful data storage
We use NFS to mount the /home directory from what used to be one of our LTSP servers. This allows users to keep all their files as they were. We decided against using NSS or Novell NCP for home directories because of input/output errors. /tmp is currently stored on the squashfs image, as are the logs. However, I think this could potentially pose a problem, since it could fill up ram. I'll probably make a new nfs share for /tmp.

Thin-Client specs:
For our new deployment, the thin-clients remain mostly the same. They still do not have local storage. When we went with LTSP clients, I wanted to go overboard on the specs. Many boast that 'Pentium II' CPUs w/32mb ram and 8mb video works great with LTSP. Since crummy small LTSP clients sold for the same as my desktops, I went the desktop option. What do they have in them?
AMD 3800 or AMDx2 4200 CPU
256 - 1GB RAM
CD/DVD ROM/RW drive
NVIDIA 6050/7150 onboard video -or- 9400GT
GigE ethernet
Turning thin-clients to the hybrid-clients only required more ram. Since ram comes a dime a dozen, this upgrade was very inexpensive. We found that systems running 256MB RAM could not display GDM, those with 512MB could run only one app at a time, and 1GB works ok. We're doing some exhaustive testing this week to see if 1GB is okay. If not, we will update the machines to 2GB. Guess how much it costs per machine to goto 2GB of ram? For 38 machines (after re-purposing the ram we have), it cost $360 from newegg. Thats right, $20/ea.

Performance improvements:
The first thing we looked at was video performance. We thought at first that nvidia 6050 cards were just crap and no better than intel or sis onboard graphics. Apparently LTSP was the problem. Tests with simple glxgears show a 200% improvement in frame rate, 9400GT got a 250% boost in performance.
Thunderbird used to render calendars poorly. So poorly that our power users had to use windows until we got this system up and running. Now? thunderbird displays every thing instantly.
QTIplot with graphs rendering. It would bring the LTSP clients to the brink of locking up (and many times it would lock up). Now? qtiplot renders perfectly.

In summary
LTSP was a great idea when the servers were less than the price difference of thin-client machines vs thick clients. All other benefits are solved with the hybrid-client solution. And since commodity computer parts are as cheap as thin-clients, there is no benefit of using LTSP.
If you want the benefits of LTSP without the drawbacks, take a look at the hybrid-client solution. Its cheaper, faster, and more reliable than thin-client computing.

Ubercart and Ecommerce once year later

2009-02-03T13:30:00.000-08:00

A little under a year ago, after drupalcon, I and some others were dismayed by some of the issues plaguing ubercart development. Flames and hurt feelings perpetuated, but both projects kept trudging along.

One year later, where have things gone? 180 degree change!

Ubercart++
So now from an outsiders perspective, the ubercart community has come together and is creating an enterprise commerce solution for drupal. Drupal 6 ubercart is *mostly* utilizing core modules that can do a better job (aka CCK, views, actions, etc), while they stay focused on making ubercart good.

The ubercart.org site is still strong on its own, but #ubercart folks are interacting more with #drupal, creating a communication conduit which is stronger than ecommerce. They haven't moved over to drupal cvs yet, but I hope that's a goal for the next year.

E-Commerce--
Although I still wish these two packages could combine, I'm seeing why ecommerce is loosing ground to ubercart. A few weeks ago two ecommerce devs were in #drupal-ubercart, finally fed up with ecommerce. They mainly pointed to the stiffness of the module maintainer, and not opening up the code for more to maintain. Now I'm not sure if this is true or not. However, my experiences were similar a few years back. While ecommerce has some good back-end features in 4.x, the development strategy is flawed because it doesn't promote more people to join and contribute.

Back to ubercart... did I leave?
I tried to convert a site back in Dec '08 to ecommerce. I kept telling myself that "the next ground up site I would do in e-commerce, and help improve it." But every time I fiddled with e-commerce, I ran into stupid bugs that reminded me why I left looking for something like ubercart.
With some new modules like uc_node_checkout, the authorize.net cim, recurring payments, etc, ubercart is very complete. Bugs submitted on ubercart.org are looked at fairly quick, and they are starting to port more contributed modules to drupal.org to integrate into the greater #drupal community.

I see really no reason why not to use ubercart anymore. The kinks have been worked out, and even with its disadvantages, ubercart is still better than any implementation of ecommerce.

But this is just my experience. Perhaps there are some e-commerce users who can talk me down ;-)

SLE11 Beta6 > OpenSUSE 11?

2008-12-10T12:17:00.001-08:00

So I'm working on the SLE beta. All I can say is WOW. in beta format its more stable than 11, has fixed most of the bugs we witnessed in 11, and I'm now contemplating if we'll release BETA software for production.

Yah yah, I know many people will say that stupid. Ok. I'd love to wait for SLE11 to be released, but thats probably 5-6 months away. And with OpenSUSE 11.1 slated for release in less than 2 weeks, I'm fairly certain the SLE betas will be stable.

Now comes working on KIWI. If I can get a SLE11 image made with kiwi, we're set. then I can deploy nbd images to each client, and they'll get their own machines. Its LTSP++.

good bye LTSP :-/

2008-12-05T13:12:00.000-08:00

I came to my current employer with the hope to try out some cool new software: the Linux Terminal Server project. In theory, with good enough equipment you should be able to deploy office grade desktops from one LTSP server.
Jump ahead 3 years. We've made great progress from the original Gentoo box running LTSP 4.1. We now have sound, local device support, and even accelerated video drivers! My thanks to Cyberorg and KIWI-LTSP project for making much of the new features available.

But alas, it just doesn't work good enough. Some issues
1) Multi-user file access isn't good enough. Novell client doesn't work, samba mount doesn't work, pam_mount, etc all either lock the server, don't display permissions correctly, or can't automount on login. We've looked at every alternative, but this issue cannot seem to be fixed.
2) Video failures. Nvidia, intel, closed source, open source drivers. doesn't matter. With nvidia, certain apps will cause the machine to hard-lock. We've tried better video cards, which seemed to fix some of the issues, but I don't think putting gaming cards in business machines is the end all fix
3) Thunderbird. Specifically, Lightning. It seems that the rendering of the calendar is display intensive, and on LTSP machines it can take upto 30 seconds to render a Month view. For some people with many calendar entries, thunderbird will become unresponsive. This is unacceptable. And was the last straw. If users can't do calendaring on their workstation, then the workstation is useless.

So what now? Using the strengths of the kiwi imaging system, I think it will be possible to deploy disk images to clients. It should only cost $20-$50 to retrofit each machine with a ~20GB hdd and 2gb of ram.
To update machines, we're planning to just deploy new images as needed.

I'm working on SLE11 right now, this should be our basis for new clients.

Any suggestions? fixes for my three issues above? I'd love to hear them.

Christmas 2008

2008-11-23T21:52:00.000-08:00

oh what shall I want for christmas?

Here are a few items I'm looking for:

*Airtools from Chicago Pnumatic:
**CP7733
**CP7830H
**CP7563

Craftsman 23" tool chest for the truck
Craftsman 26 in. 6-Drawer Ball-Bearing Chest for the shop
Slide out cabinent for the rear of my 4runner. Since my truck is cut in the rear, here are the dimensions: D28"xW40"... normal depth is about 36".
here is writeup of something similar
LED rock light kit
GlobalSAT BT-359C (or BT-359W) Bluetooth GPS Reciever
LED Headlamp (like this example.. don't really care what brand, just needs to be durable and water resistant/proofish)
DeWalt grinder (or similar)
Dell Inspiron9 Laptop (Its a mini-PC, good to go with the GPS..) but its ~$479
Nice pants, slacks are good.. 32x32 dimensions.

Other cool things:
Gift card for McLendons
Gift card for Hardware Sales
Bawls Energy drink

Nvidia 177.80 RPMs for OpenSUSE 11.0

2008-10-29T23:11:00.000-07:00

After going through the novell bugzilla, and grabbing the source files from the Suse build service, I created the RPMs for OpenSUSE 11.0

If you wish to skip how I built the drivers, click below for the RPMs.

Requires OpenSUSE 11.0 with 2.6.25.18-0.2-default, pae, or lockdep kernel
edit: updated to work with BFG 9400GT cards
nvidia-gfxG02-kmp-default-177.80_2.6.25.18_0.2-3.i586.rpm
nvidia-gfxG02-kmp-pae-177.80_2.6.25.18_0.2-3.i586.rpm
nvidia-gfxG02-kmp-lockdep-177.80_2.6.25.18_0.2-3.i586.rpm
x11-video-nvidiaG02-177.80-2.i586.rpm

To build from source:
osc checkout X11:Drivers:Video/nvidia-gfxG02
copy the contents of X11:Drivers:Video/nvidia-gfxG02 into /usr/src/packages/SOURCES
rpmbuild -ba nvidia-gfxG02.spec
rpmbuild -ba x11-video-nvidiaG02.spec
rpm -qa | grep nvidia ==> if you currently have nvidia-G01 installed
rpm -e (nvidia rpms)
rpm -Uvh /usr/src/packages/{arch}/nvidia-gfxG02-kmp-{flavor}-177.80-{kernel_version}_{rpmversion}.i586.rpm x11-video-nvidiaG02-177.80-{rpmversion}.i586.rpm
depmod -a
modprobe nvidia

If you're using a newer video card you might find that the driver cannot find your device. In that case, you can try my alternative .spec file. (x11-video-nvidiaG02.spec and rpm doesn't change)
nvidia-gfxG02-newdevs.spec
save this file and copy it to /usr/src/packages/SOURCES
rpmbuild -ba nvidia-gfxG02-newdevs.spec
rpm -e nvidia-gfxG02-kmp-{flavor}
rpm -Uvh /usr/src/packages/{arch}/nvidia-gfxG02-kmp-{flavor}-177.80_2.6.25.18_0.2-2.i586.rpm
depmod -a
modprobe nvidia

If you're having this problem, submit a comment with your hardware info by pasting the output of `hwinfo --gfxcard` or submit it to the bugzilla tracker.

Clarifying Ubercart and Ecommerce

2008-09-26T14:03:00.001-07:00

In the past few weeks I've had people ask me more about my position on ubercart and ecommerce. Probably due to Stephanie's post about ubercart in her blog. Anywho, I'm creating this post to update what I've observed as an outsider, developing on ubercart and the movements being done to address concerns in my previous blog post.

1) Ubercart development process
I still believe work needs to be done in this field. Case in point: Ubercart 1.4
http://www.ubercart.org/news/sep-17-2008/ubercart_14_released
If you compare this to drupal development, you'd never see these type of issues when updating drupal 5.9 to 5.10. Why? because they have a good case tracker, people do bug testing on HEAD before it gets released, and most importantly NO FEATURE changes! Ubercart 1.5 is planned to be released this week to fix issues caused in 1.4. If this was drupal development, 1.4 would have never been released, instead these issues would be tested first on head and released properly.

That said, there is some good work being done to make the software better, and for many users ubercart does what they need. They're getting features built and released, keeping the community happy.

Another big plus to the ubercart devs: Features are starting to be rolled into the 6.x version, and not 1.0. I think we'll start seeing a 'real' stable version of ubercart soon.

2) Ubercart vs Ec.. which one?
I'm recommending ubercart to most users. It has the most complete system out there. It might be monolithic, but until eC is able to bring a similar UI, ubercart is where its at. However, if you're looking for a framework for your own ecommerce solution, or you only need a few ecommerce functions, the eC package may work better for you.

3) Drupal community interaction
Its looking like Ubercart will end up splintering the drupal community whether I or others like it or not. People doing ecommerce usually flock to ubercart.org and aren't seen in #drupal as much. People who develop on d.o aren't 'ever' seen in ubercart.org's page.
However, things are starting to change here as well. Issues are being tracked in the d.o issue tracker, and CVS is starting to be used more in conjunction with bzr. Although there is still work to be done here. More beginner conversations are taking place on d.o regarding ecommerce, and are being answered here.

So in summary: Use ubercart, but there are some things to watch for. Upgrading ubercart is not as simple as drupal, because of the changes made between 'stable' versions. The dev team is supportive and understands the fundamentals of ecommerce, although they still need to learn more about software and large project design. There are many companies using ubercart. If you know the pitfalls and can work around them, it'll work great for you.

Updating OpenSUSE 11 round 2

2008-09-18T10:44:00.000-07:00

I've been hacking on my devbox here at work. I installed the most recent updates, including kernel 2.6.25.16-0.1-pae

First issue: Novell login scripts don't work. Someone commented about this in my previous OpenSUSE blog. Verified, it does the same thing to me.

second issue: duplicate icons on desktop. The easiest way to delete the KDE icon is by chmod 700 on kde-thumbs. This will cause the icon to not appear

Third issue: larger PDFs run incredibly slow. Got a suggestion to run Ocular. Its a KDE4 app, and we're a Gnome shop. However, it seems to work pretty well.

I'm using the following repositories for OpenSUSE 11:
openSUSE - OpenOffice.org
KDE:KDE4:STABLE:desktop
VideoLan Repository
openSUSE BuildService - GNOME:Community
Packman Repository
openSUSE BuildService - Mozilla
server.ltsp
openSUSE BuildService - GNOME:Stable
openSUSE BuildService - X11:XGL
openSUSE Education
openSUSE-11.0-Updates
Main Repository (OSS)
Main Repository (NON-OSS)
KDE:KDE4:STABLE:Extra-Apps
NVIDIA Repository
openSUSE:Tools

I'm using easy-ltsp now to build my images. I don't use the prebuilts because I have some custom code. Using the post at http://en.opensuse.org/LTSP/Localapps I've been able to build nvidia support into my images.

Unfortunately I could not get the updates to work within the build images, so I have to copy updated RPMs manually into the 'extra-packages' folder within LTSP.

Luckily bootsplash has been fixed in recent versions, so has errors on startup. Users now can see the opensuse splash screen all the way upto login.

The last thing I need to do today is test LUM support on the machine. If it works properly, I'll deploy it out to one production box, then to the other!

LUM users dropping group?

2008-08-05T10:46:00.000-07:00

We've been running OpenSUSE 11 with LUM for about a month now, and we're running into an issue with users randomly loosing their primary group.

You type the following:
'id user'
returns:
uid=1044(user) gid=601(Domain Users)
It SHUOLD be returning
uid=1044(user) gid=601(Domain Users) groups=601(Domain Users)

I thought at first there might be a Identity Manager issue, where Windows is causing the user to loose group information. So I disabled synchronization, but alas -- no luck. While the error seems to be happening less now, its still happening.

The only way to fix the issue I've found so far is to do the following:
open up iManager
modify the user
delete the group they belong to which is a 'primary group'
add the group they should have as a primary group
and it shows up again

Live from OSCON!

2008-07-19T00:38:00.001-07:00

So all this week, I'll try to update folks on how things are moving the LTSP and Drupal world.

First things first though. setting up the makeshift lab in the hotel room. CHECK.

If you wanna see our test webcam? Its up!

it reloads every 4 seconds. I need to get longer ethernet cables so I can put it in a better spot.

Live from OSCON!

2008-07-19T00:38:00.000-07:00

So all this week, I'll try to update folks on how things are moving the LTSP and Drupal world.

First things first though. setting up the makeshift lab in the hotel room. CHECK.

If you wanna see our test webcam, go here: http://mail.asemblon.com/webcam/oscon.html
it reloads every 4 seconds. I need to get longer ethernet cables so I can put it in a better spot.

OSCON 2008

2008-07-18T03:25:00.000-07:00

its 3:25am, ugh and I'm still up. Went to see Batman tonight. GREAT, err EPIC movie!

So far OSCON is looking like this:
Leave tomorrow, Friday the 18th -- going to talk with the postgreSQL group at their meeting on sunday. Perhaps goto the beach Saturday.
Monday, Tuesday working on drupal stuff -- possibly some LTSP, depends on who will be there.
Wednesday I'm going to try to meet up with Greg KH, either on the expo hall or at the Opensuse meeting at 4:30PM.
Thursday night is FOSCON, and Josh + Me are supposed to be competing in a rapid web development thingy. It should be fun to show off Drupal!
Friday starts the LTSP conference, going into saturday. I'll be around hopefully most of friday, and try to be there saturday morning-midday, before heading back home saturday evening.

If you want to stalk me, here is my (tentative) schedule:
http://en.oreilly.com/oscon2008/public/schedule/share/8150fc1b3e34c209fe837a3e0d3fd926

At the linuxfest northwest booth, we will have our production/test LTSP server, running OpenSUSE 11.0 and the Novell client. If I can get the VPN running correctly, I hope to show it off connecting upto eDirectory.

My main goals at OSCON this year is to..
1) talk to some novell/opensuse devs and get an idea where Novell is on supporting multi-user systems such as LTSP
2) Kick ass and take names with drupal development
3) Find neat new tools we can implement within our technology structure
4) Try to squash a few annoying usability problems with our current LTSP system.
5) Find at least one person who uses tinyERP. (Good luck with that!)

If you're looking for me, I'll be wearing a computer t-shirt. oh and have a name tag on. umm I guess that describes 90% of us. I usually lurk around the linuxfest booth.. hmm that probably doesn't help much either. Okay well I look like this... I'm the young one, bottom right:

Novell Client 2.0 after a few days... =(

2008-07-16T17:32:00.000-07:00

After my previous post about running Novell Client on SuSe, we upgraded all of our SLED10 users to OpenSUSE with success at first. However this morning someone called me up and said:

"Hey Jakob, Public isn't showing up"

Uh oh. We saw this the second day after implementation, but the user just had the public icon behind another icon. This time I was not so lucky. My first hunch was novfs. Oh man, good guess:

drwxrwxrwx 1 root      root   0 Jul  9 19:10 .
drwxr-xr-x 6 root      root 144 May  8  2007 ..
drwxrwxrwx 1 root      root   0 Jul  9 19:10 .Servers
drwxrwxrwx 1 root      root   0 Jul  9 19:10 .Trees
drwxr-xr-x 1 admin     root   0 Jul 15 09:33 admin
drwxr-xr-x 1 ctimmon   root   0 Jul 15 09:28 ctimmon
d????????? ? ?         ?      ?            ? dhson
drwxr-xr-x 1 root      root   0 Jul 14 18:54 djanss
drwxr-xr-x 1 eesrez    root   0 Jul 15 07:43 eesrez
d????????? ? ?         ?      ?            ? esagh
drwxr-xr-x 1 gblte     root   0 Jul 15 09:28 gblte
drwxr-xr-x 1 gdvis     root   0 Jul 14 16:57 gdvis
drwxr-xr-x 1 mbokl     root   0 Jul 15 07:27 mbokl
d????????? ? ?         ?      ?            ? mdrak
drwxr-xr-x 1 mnaemi    root   0 Jul 15 08:16 mnaemi
drwxr-xr-x 1 tbkr      root   0 Jul 15 08:00 tbkr

Wtf? ??? ? ? ?? ? ? there are Q's all over the place, meaning I believe that whatever links user/permissions to the folder has gone away. Digging back to my C experience (which is little and lasted all of one semester at PLU) I figured there is probably a buffer overflow or some max setting that's being hit. When we had our 5 test users, we ran this for 2 weeks without a problem. We put on 19 users, and once we hit that magical number, we start getting very odd problems.

So the quick fix is to only have 15-18 users logged in per machine. For our company, thats okay... for now. But this is a major bug. Who knows what security problems exist within this module.

I opened up a bug on Novell's website: https://bugzilla.novell.com/show_bug.cgi?id=409308

My thinking now is that novfs really should become a FUSE module, which could work with the novell client to restore network maps. I'm hoping someone from novell sees these blog posts and helps out.

Am I the only one who runs multiple users per machine?

Novell Client 2.0 works in OpenSUSE11!

2008-07-09T15:31:00.000-07:00

I'm not sure if I'm the only one who uses the novell client to log multiple users in on one machine, but bug reports sure don't show a whole lot of progress.

However! today, I think I have a patch that will fix the NWdrivebase issue seen here:
http://www.fcdnet.org/japerry/2008/05/novfs-ownership-woes.html

Basically there is some problem with random users getting assigned to the root directory that mounts servers for the novell client. Since I couldn't find where the user failure was, I decided it'd be good enough to change permissions. The sub folder isn't affected, so security is retained.

Patch here: http://fcdnet.org/japerry/novfs-permissions.patch

To apply this patch, go into fs/novfs in the kernel-source. You'll have to rebuild the kernel (but don't have to make install it) but just copy /proc/config.gz into your src directory, unzip and rename it .config. Make sure the .config includes NOVFS=m. Apply the novfs patch included here, type in make, and wait for a bit. When finished copy the novfs.ko into /lib/modules/KERNELNAME/kernel/fs/novfs/ -- if you want, backup the current novfs.ko file in the folder.
This should change the permissions from 0700 to 0777, which will allow the user to delete the directory upon logout. You'll notice that once inside /var/opt/novell/nclmnt/USERNAME that the Servers are all 0700 owned by the correct user.

Drupal 6 + Ubercart sandbox

2008-06-04T16:23:00.000-07:00

So coming off the heels of the last post, I'm still working on ubercart. Blah. Many people use ubercart, and drupal 6 support is badly needed.

That said, I've done quite a bit of work on the matter, and decided to post my sandbox here. Before you download, some caveats:
1) This is pre-alpha code -- ubercart v6 is only partially done and needs more development work done to it
2) it'd be better if I submitted a bunch of patches, but someone wanted a working system quickly--this is the fastest way to get up to speed.
3) this is RC4 code. I'm working to do a bzr update to 1.0, I'll update the tar.gz file as soon as thats finished.
4) This module comes with coder, devel, ubercart, rules, workflow_ng, ubrowser, tapir all with d6 .info and .install files and partial module support.

Download Here

Instructions:
Download, and extract to a directory accessible by your webserver. I use MAMP, which creates a nice little webserver/db for my mac.
Import the drupal6.sql file into mysql (mysql only right now, not postgres!)
Change settings.php to reflect the relevant paths for your installation.
My sandbox was just drupal6 (eg: http://drupal6/)
Bring up firefox and see if the site comes up. change files location to something relevant to your file system.

And that should be it! Email me or find me in irc.freenode.net: #drupal-ubercart, #drupal-dev, etc

Note: edit Sept 1st, 2008: ubercart has been getting towards D6 quickly -- support.ubercart.org or bazaar.ubercart.org/drupal6

novfs ownership woes - NwSetDriveBase2:8804

2008-05-21T10:08:00.000-07:00

Something that has really been bugging me is a pest error called NwSetDriveBase2:8804
This error has showed up since SLED10 and persists into SLED10SP1, OpenSUSE 10.3, and OpenSuSE 11 Beta3.

Someone has posted another related issue with this error, but it seems to be unrelated to what I'm facing.
http://forums.novell.com/novell-product-support-forums/open-enterprise-server/oes-platform-independent/oes-client-linux/143685-nwsetdrivebase2-error-mapping-drives.html

I've found two causes to this error. One is an easy fix, the other is the more annoying one...
1) Invalid/Dead symlinks to /var/opt/novell/nclmnt -- when ncl tries to map the drives it sees the file already exists and throws this error. Novell issued a KB on this:
http://www.novell.com/support/viewContent.do?externalId=3810237&sliceId=1

2) Bad ownership on /var/opt/novell/nclmnt -- Example 'ls -la'
drwxrwxrwx 1 root root 0 May 7 17:51 .Servers
drwxrwxrwx 1 root root 0 May 7 17:51 .Trees
drwx------ 1 adfdgel root 0 May 21 07:27 adfdgel
drwx------ 1 fdfdard root 0 May 21 09:05 fdfdard
drwx------ 1 dhgfdon root 0 May 21 10:14 dhgfdon
drwx------ 1 nobody root 0 May 12 18:16 ghdfsen
drwx------ 1 root root 0 May 16 08:26 rsiueon
drwx------ 1 nobody root 0 May 19 18:16 tqkher
**usernames mangled for security**

You'll see above that the first 3 users are working correctly. They have ownership of their directories and are logged in. Users 4 and 6 are not logged in, so their folders are owned by nobody. Thats fine. The problem is with user number 5. rsiueon IS logged in, but his directory is owned by root, therefore he cannot map his drives. However, it seems that if this user goes into the .Servers directory, he can see all the shares just fine.

The only way to fix this is by rebooting. We've tried to log the user on and off multiple times, with no success. I'm not sure whats causing this to happen randomly, but its by far our biggest day-to-day issue with the novell client.

LTSP / KIWI / OpenSUSE 11 / Novell eDirectory

2008-05-20T14:33:00.001-07:00

Below is the checklist I compiled for our new OpenSUSE 11 KIWI-LTSP server, which pushes out images for ~25 clients. This list isn't totally complete, and somewhat customized for our company. However, it should get most people where they need to be if they're using OpenSUSE with Novell servers.

Updated -- Added info about the NWsetdrivebase2 error here

I'll add our custom pam_scripts if demand warrants it. Let me know if you're having trouble finding any of the packages to install, and need a copy.

* install opensuse 11 DVD
-- boot partition /dev/sda1 - 100M
-- root /dev/sda5 - 70G
-- var /dev/sda6 - 10G
-- swap /dev/sda7 2G
* Install gnome, development packages, KDE3 libs

Checklist:
kiwi-ltsp
* one-click install for SuSE 11
* copy our kiwi-ltsp-backup to /srv
* make sure nbd xinetd is not running (disabled = yes)
* after kiwi one-click setup is done, change the nbd device pointer to the right image
* set uid lbmount +s
* chkconfig nbd-server on
* copy over rpms required for nvidia graphics

Novell Client / Pam script
* make sure kdelibs3 and gtk are installed
* download / mount the NovellSUSE10.2 iso (not SLED10SP1)
* copy the contents to /opt/ncldisk
* edit ncl_install to get rid of novfs (but not novfsd)
* link libbfd to proper library in /usr/lib/libbfd-?
* rpm --nodeps novell-xtier-base
* rpm --nodeps novell-novfsd
*edit /etc/init.d/novfsd and eliminate mk_novfs and the cd command before it
* ncl_install
* rpm --nodeps novell-client
* yast2 --> network services --> novell client
Install PAM
* download novell-NLDAPbase and NLDAPsdk
* rpm -Uvh novell-NLDAP*
* rpm -Uvh novell-lum-2.2.0.14-3.7.i586.rpm
* rpm -Uvh yast2-linux-user-mgmt-2.13.1-0.8.noarch.rpm
* download newest pam_script and compile for system
* unpack the kioskserv /etc/pam.d scripts
* edit /etc/nssswitch for passwd and group should say: compat nam
* unpack the kioskserv /etc/security scripts
* run namconfig add -a cn=MYDOMAIN,o=NOVELL,o=com -p PASSWORD -r cn=MYDOMAIN,o=NOVELL,o=com -w ou=UNIXSYSTEMS,cn=MYDOMAIN,o=NOVELL,o=com -S SERVERIP:389 -l 636
* namconfig -k
* /etc/init.d/nscd stop
* chkconfig nscd off
* /etc/init.d/namcd restart
restart server
when restarted, test by typing: 'id (NDS USER)' and getent password (NDS USER)
---below we link our /home directory to our nfs server---
note: as of this checklist, we cannot figure out a way for user's home directories to be mounted as NSS shares from eDirectory without any root directory first. That and we've had bugs where novell login will fail. So our other option is to create a home directory on our file server for thin-clients, then link their documents and a public directory to their nss shares through the login script.
To do this, perform the following:
add the /home directory to fstab
SERVERIP:/home /home nfs defaults 1 2
mount -t nfs SERVERIP:/home /home
check home nss: cd /home && ls -al --> should output usernames not uid/gids
copy over /etc/hosts file

Christmas 2007

2007-11-01T21:46:00.000-07:00

Okay so here comes the christmas list of 2007:

Couch (three cushion, with the sides as recliners)
Well constructed dining table (so no ikea, walmart, etc) w/ chairs
a nice plug-in drill (not cordless) with 1/2" drill input size (not 3/8s)
Jig Saw to cut metal
large car jack (3ton+)
king coils
ski tickets (Crystal, Baker, Big White, Stevens)
220Watt TIG Welder
Steel Pipe Bender
VHF Receiver/Transmitter
GPS unit that works with OSX
KDC-MP535U CD Car Stereo
SAE Ratcheting Wrenches (Not Metric)
Self-Cleaning Shaver
KC HiLights or similar 4x4 lights (bright, not running lights --something similar to my Jeep lights) -- for reference: http://www.autotrucktoys.com/liberty/Mopar-OEM-Jeep-Liberty-Light-Bar-Kit-02-07-P6952C1820.aspx
--note: they DO NOT cost what that website says they do
--for my toyota -- so something that fits a Toyota truck 79-88

Birthday 2007!

2007-02-05T22:33:00.000-08:00

Heh so this Birthday will be easier than 'ever! There are a whole bunch of little 4x4 parts I'm in need of, and they add up. So here are a list of supplies I'm looking for:

*mounts may or may not be Jeep Liberty (KJ) specific*
Tool Box
Fire Extengiuser Mount
D-Cell Mag-Lite Mount
Air Pressure Gadge / Tire valve tool
CB Antenna and Mount
Head-Mounted LED light
Jeep Slush Mats:

I'll update this as needed. but those things would be really cool.

Veggie 300SD

2006-06-01T18:37:00.000-07:00

In April, my good long friend Bovie turned me on to a new idea that is making a movement into becoming free from the oil economy we're living in--Waste Vegetable Oil (WVO) in cars. I vowed to never pay over $2.00/gal for gas back in 2003, and had to reclutantly eat my words. Well I said it again for $3.00/gal back in Sept 2005, managing to pay 2.72 at the highest peak. Well this time, no Katrina, no opec, just the plain facts that as long as people demand oil, the companies providing it will jack up thier prices. In April I noticed that 3.00/gal was coming back soon, and I really did not want to eat my words again. So Thanks to bovie, and the guys over at http://www.greaseon.com/ , I started the search.

Choosing your Diesel

So I knew I wanted a mercedes. Why? well first, Kent Bergsma, the mechanic who helped convert bovie's 240D, said that he has only converted mercedes cars, and although it can be done to any diesel, it wouldn't be as easy. I had a bias already agains Volkswagons because I've had not one, but two roomates that have had Jetta's die on them. So it looked like an older VW was out of the picture. Didn't want a truck, and an old rare diesel caddy didn't look like a good idea either. So off to ebay and craigslist.

Craigslist and Ebay

I started looking through these sites to see what I could find. Most newer models, 1995 and above were WELL over 8000k, unless they had some major problems with them. Craigslist vehicles were even more expensive. I called a guy in Portland regarding a 1995 S350 Diesel. Its a nice car, but he wanted 11,200 for it. He suggested that if I'm just starting out, that I should go for an older model, and referenced me to an older 1983 300SD Turbo diesel. So I took the train down and looked at it. The car didn't look in that bad of shape, and it only (yes only) had 186,000 miles on it. So despite is 'issues' I bought it off him, $2600.

The car

So the car is a 1983 Mercedes 300SD Turbo Diesel. It rides fairly well, and was a perfect candidate for conversion. Actually looking back, buying this car instead of the newer W140 or W210 models was actually a good idea. The 300SD is based on a w126 chasis, which use very little electronics to make the car run. Kent advised me that the best models are the w123, w124, and the w126. the w140s had some pretty major electrical issues, I guess spawning some recalls in 1995. The best model he suggested would be a 90-93 300D, 2.5 diesel. They're built on the w124 chasis. Perhaps that will be my next car. This car is getting from 27-30mpg on diesel, and I filled it up with Biodiesel at Laurelhurst Oil in the U-District in Seattle.

The Fuel

Secure your fuel source BEFORE BUYING A CONVERSION KIT! I put this here because its great to buy a mercedes and run it on biodiesel. However, before you decide to convert it to run 100% WVO, you need to find a place to give you oil, or else you've just spent $1000 to heat up already thin bio-diesel.

Difference between Biodiesel and WVO

Biodisel is thin, WVO is thick, very thick

WVO can be put into a converted vehicle just by filtering out the crud, Biodiesel must be refined

WVO needs to be heated up to ~150F in order for it to not f&&K up your injectors. You can run your car at a lower temp, but it'll mess it up. without any heat, your car won't start.

Biodiesel runs without any conversion, although its best to get your fuel hoses replaced and your tank cleaned

When you hear people say they are using vegetable oil for biodiesel, that could very well be true. However, since biodiesel requires other ingredients to make it work, you'll end up paying per gallon, while WVO, after the initial costs, will virtually be free, not counting hours to pick the stuff up.

Converting the car

We'll get into some nitty gritty stuff here, hopefully by the end of this post, you'll have a good idea what it means to convert a car to run on WVO.

First: Choosing a kit

At first, bovie told me to go with Kent's system. I looked around like a savvy shopper for other systems, noticing that other systems are cheaper and 'looked' like they could do the same thing for me. However, don't be fooled. Systems from greasecar, frybrid, and greasel can be nice, however the 'econo-systems' lack some very important parts, which can usually result in reduced performance. The other thing to note is that the greaseon kit took Bovie and I about 18 hours to install. Bovie used a combo of greasecar and greasel systems, which took weeks. Since greasecar assembles its own brackets for both the front and rear of your car, you just need to hook it up, and not assemble it (for the most part). So once I purchased the SVC-1500, from greaseon, I awaited the boxes to start working on it.
(note: I must admit that we were able to use Kent's lift for all of our underbody work, which we couldn't have done otherwise. Also we had his ear whenever we hit a bump, which clarified issues that might not have been as easy on our own. However, this reason alone was not why I choose greaseon over other systems)

Second: Reading DIRECTIONS!!!

So I must admit, we somewhat learned this the hard way. Another advantage of the greaseon system is its quite verbose documentation. If you're a good reader, this documentation will result in a flawless installation. However, if you have CRS (Can't remember shit) and don't do good studying, you can end up borking some parts.
At this point I should note that you should also keep the parts in boxes that come with the system separate. We sorta poured everything together, which, without Kent's help would have been disasterous. The lines are cut for specific uses, but if you blend them, it can be hard for the novice to know what goes where. That said, onto the installation....

Third: Assembling the fuel lines

There is one thing you must asemble yourself, the fuel lines from the rear to the front of the car. This is because if Kent did it, the shipping costs would be out the window. Plus, with a second person, its pretty easy to asemble. We had it done within an episode of the simpsons.

Make sure you get the tape tightly around the hoses, and make sure to read the directions about wire alignment!!! IF you don't, you'll end up with some wires that are WAY too long on one end, while other hoses aren't long enough. We had to do some splicing after missing this little, but important detail.

Fourth: Asembling the FSB

After the fuel lines are in the system, you'll want to mount the FSB to the back, in my w126, it fit nicely next to the spare tire, just like it was designed to. Try to dril the hose line going to your fuel line before you get your FSB mounted, that way its not as hard to drill. Also make sure you have ample clearance between the pump and the top of the spare tire well.

Fifth: Replacing the Fuel tank fitting

This is where a lift comes in REALLY handy. I'm not sure how many more hours it would have taken otherwise. the tank outlet is located near the rear-right wheel well, you'll notice two lines, one for output, and one for return. Its useful to have a near empty tank when performing this procedure, because you need to empty your tank and dry off the fitting. Let the tank drain out. Once completed, you'll need to use a vise or other tool over the fitting to get it off. Since we're replacing the smaller fitting with a larger one, you'll need to unscrew the nipple part from the tank part, which will break the seal. Install the new fitting and put some provided epoxy. Or if you're like us, get some quick setting 30min epoxy and you're on your way.

Sometimes it can be difficult putting the fitting back on because you'll need to tighten down the black fitting, which might be butted up next to the fuel tank lip, making it impossible for a ratchet to get in there. We had to use a hammer and punch to get it tighened.

Sixth: Install the fuel line

You'll most definitely need a good drill for this process. Also is nice to have a lift. Drill a pilot hole where you want the line to go. It should be coming through the left side of the spare tire wheel-well. Once that is all drilled through, bring the fuel lines through and match the individual ends up to where the fsb is. If you did this right (We did not), you'll have a coolant line that is long enough to go around the outside of the well, and the other lines go through the inside, including wiring.
Next, take the fuel line and start tie wrapping it next to the old fuel metal lines on the car. you don't want to tighten them quite yet because you might need to move the lines one way or another. Once the lines are fitted to the back, you can start bringing them to the front. Again, a lift is highly recommended. Push the fuel lines near the steering column. You'll want to try to get it tied off so as it comes into the engine compartment, it doesn't rub on anything.

Seven: Install the VFM

Here are the nuts and bolts of the system, sorta. The VFM is the brain of the electronics, switching the system via the switches in the car (later), providing the last fuel filter before the oil hits the injection system, and lastly the veg-therm. Some other kits have a hot veg-therm, and if you don't use greaseon, make sure the other kit provides you with one. This is what provides the magic at startup so you don't have to use Diesel to get the car running. It also will lengthen the life of your injectors because the oil is really hot so it won't gum them up. In theory you could just use the coolant (which I used once, when a relay died on me) being supplied to the back of the car, which is warm enough to heat the oil in the front, but you will damage your car after prolonged use without the veg-therm.

Seven a: Mounting the VFM

In my w126 there was a wierd bracket in our mounting way, so we just removed it.
if you don't have this thing, then just mount it down, drill through and tighten.

Seven b: Wiring the VFM

If you have two people, tell the second person to move to the next part. In our case, Bovie did the hose plumbing and I did the wiring. Match the colors suggested in the book, there will be an extra wire that could be used in the future.

You'll probably be perplexed as to why the fitting on your fuse isn't the same as the fuse box. Take some vises or needle nose pliers and bend open the fitting to fit over the fuses. If you're lucky like me and have a fuse key, find the wipers and put it over that fuse.

Seven c: Plumbing the VFM

IMPORTANT: Follow instructions VERY carefully here! You'll be cutting hoses to length, and if you mess up, it can be difficult to recover without many splices! Also, have a spill container below your car, or else you'll have a nasty coolant and fuel spill to clean up afterwards.. and note to the note: get extra coolant if you're picky like that. If you're like us, you'll end up loosing half of it
First, find your coolant lines, they will be up near the wheelbox and down in the front--just on top of the alternator. You'll use the big 'T' connectors and cut a splice. Both of these are a bitch to splice because its hard to get to them and they don't give you much slack.

Also, make sure you put your front coolant line behind the air intake and tie it off to the engine block. It only took 5 days for my coolant line to get cut by the fan! Once those wires are spliced, start going for the fuel pump.

Seven d: Plumbing the VFM: Fuel Lines

If you're using a one tank system, you will just be plugging the outlet line into the one bi-valve included with the kit. For two tank systems, like mine, you'll have two bi-valves.

Since the fuel intake line isn't being used anymore, we can use that for our second tank, otherwise it won't be used at all. The outlet line will always be going into the main tank, so if you have a Second tank, MAKE SURE YOU DON'T HAVE RETURN FUEL SWITCH ON!!!! or else your aux tank will be gone in a matter of minutes! Back to the single tank instalation.
For a one tank system, our new fuel line will go right into our fuel filter. the aux fuel system does it a different way, mentioned below. Make sure that the fuel filter has the supplied O-ring on it:

Mine didn't come with one, and when we turned the system on, veggie went squirting everywhere, causing me to have to replace my belts because they became greasy.
The filter will go into the veg-therm, the veg-therm into the fuel Bosch pump, then straight into the Injector. The outlet of the injector goes back into our loop, and is repeated over again. With a 2 tank system, the aux tank will use the regular fuel filter that we just bypassed with the veggie wires.

now this may seem very confusing, any perhaps you just skipped over the previous paragraph--read the book--it'll describe it better.

Once plumbing is done, good--the hard part is over.

Seven e: Installing the Fuse in the VFM

The 3 wires coming into and going out of the VFM can be somewhat tight, make sure you have the groove right (read manual) and you should get a good fit. Don't install the fuse quite yet, but first make sure you have the black 12gauge wire going to your battery. (which feeds from the bottom of the fuse, the 3 wires are on top). ITs recommended to get an Optima battery, I purchased a yellow top one for $85 on craigslist.

Eight: Wiring switches

If you have a one tank fuel system, you'll only have one switch--this is the return fuel switch. Which, unless you need to purge the system, you'll never use this switch. And like I mentioned before, if you have a two-tank system, be careful with this switch or else your second tank will run out of fuel VERY quickly!

Although blurry, I brought Six wires for my two switches up under the dash and through to the center console. I decided to put the switches to the lower right of the steering column, so they are out of the way, but easy to reach as well. Read the instructions on the switches as to how to wire them, and perhaps label the switches to know which one goes to what--unless you remember colors really well. (Mine: amber--return fuel, red--aux tank)

Nine: Aux Fuel Tank

Aux fuel tanks can range anywhere from $20 to $500, depending on size, quality and fit. I choose the cheap marine type because it was easy to install, and was really just a proof of concept.

You'll want to run the fuel line outside of the spare wheel well, I found a spot right to the left of the muffler, which worked perfectly. You should get a C-clamp to screw down the hose so it won't hit the muffler. PLug the aux tank into your original fuel line, and you're good to go in the back.

In the front, you'll have the second bi-valve, mentioned above. Put the fuel line into here, with your veggie fuel line. The output from this valve will then go into your fuel filter.

Purge!

So we had an empty fuel tank, and an empty aux tank. Thats a whole lot of air in the system. You have to manually pump fuel up to the front from the aux tank, although with the SVC-1500, the fuel pump will get you fuel really quick. Sometimes the injectors need to be messed with to work, but luckily we didn't have this problem so about 10mins of pumping and testing and we had fuel to the system

Turn on!

If everything is set and working, you should be able to key on the system. Test first by turning it to the 1st position on the ignition. HAve someone outside touch the veg-therm when you turn it on to make sure it comes up. If so, wait about 30secs and then turn it to glow, and start. If everything is right, congrats you have a veggie car!

Filtering the fuel

So once we had the car running, we needed fuel. I'm getting mine from a nice teriyaki place in Kirkland. Its really clean and they produce I think enough for me to run on indefinitely.

Make sure you have gloves that are good. This stuff is messy and doesn't come off easily.

Bag filters suck. Find $400 and get a pumping system, especially if you are going in with someone else to get a few cars to filter. Bags just suck royally.
I forgot to snap a photo of some bad veggie, but here is what to avoid--Hydrogenated anything oil. Or shortening. This means that McD's, Taco Bell, *insert fast food crap restaurant here* do not work. I have a guess that applebees also uses crap oil. WWU had about 60 gallons of oil available, but it was all hydrogenated as well. You want to stick to gourmet restaurants, chinese places, and teriyaki bars. Mostly everything else is bad. Greek might work too.

Once you get some oil, and its filtered, pour it into the car. Yet another reason why bad filtering sucks. You need at least two people to get the oil in with 5-gallon buckets and a funnel. So go buy a pump system, trust me its worth it.

Complete!!

So there ya go. I'm now %98 free from the oil economy. Sure every 3000miles or so I need to use some petrol-oil for my car, but every day driving is now free. or at least more free than it was before. And within 10-12 months (perhaps less if I go on road trips this summer) it will have paid itself off financially. Not to mention I now know how to work on my 300SD!

Christmas Wish List

2004-12-06T20:32:00.000-08:00

Okay so its taken some time to compile this, but these are things I've been eying lately.. in order of want/need: (Click to goto a place to buy it, or put the product in an Ebay search)

1) Apple Bluetooth Keyboard
2) GOOD Ski Jacket (I'm still trying to figure out what this is.. I'll know this weekend)
3) Neuros Mp3 Player
4) Sonicare Toothbrush
5) Normal day thermal gloves
6) 20Pk NiMH Batteries (2200+)
7) 200+ CD case (2 actually)
8) Shaver (Nix this.. mine is working OK)
9) Ski Lift Tickets (Crystal, Baker, or Stevens)
10) ooh a Labeler would be coo

Thats about it on my list at the moment...

Vancouver Firefox Release Party -- EggOn .1!

2004-11-21T18:05:00.000-08:00

After seeing the release of Mozilla Firefox, I decided that it was imparative to goto some release party. However, Seattle, being the Microsoft capitol, didn't really have much in the way of a party. Vancouver on the otherhand it looked like was going to have upto 50 people! Well, about a third showed up, but it was still fun... Here is the group

It was funny though getting there, I could have sworn we were at a Mac Convention of some sort. The only person with a laptop was Luke, who had a 17" Powermac. it was VERY nice.... Later on, another person came bearing a laptop as well. Yes, you guessed it, a mac.

Since the crowd was only about 20 or so, after the 2 girls left (Yah we didn't have the 'Ontario' turn out we expected) we started getting desperate for fun. We had wireless in the pub, so we decided that it was time to Wikalong some ASCII pr0n....

However, our waitress wasn't too happy about that... so we turned it off and went back to drinking... which, might I add, she liked much more! We tried explaining why Firefox is so much better than that other browser, but I don't know if she figured it out or not.
Around midnight or so we decided to head out and grab pizza. Vancouver is known for its cheap pizza by-the-slice. So for $1.25 a piece, we got some grub and started talking...
"What should we do now?"
"hummm.. Lets program a Firefox Extension!!!"
So we decided to goto Kalien's house and start coding. Its 12:30am.....

We arrive, and come up with these ideas: IrcAlong (like Wikialong, but live chat)
BlinkOTron (randomly turn HTML elements into BLINK)
FuckOffLeavemeAlone (make links go display:none when hovered over)
ClickOnHover (when you hover over a link, it goes to the page)
And... Eggon.. and Egg Timer for Firefox!

We start the planning process, grabing sheets of paper and a project board. Once we figured out what frivilous project we wanted to do, we got started.... We choose Eggon!
First thing.. we needed supplies. So we went out and got some eggs and pop. After we got back, Dave and I did research on how long you should boil an egg. We came up with:

4 Mins for Soft Boiled
8 Mins for Medium Boiled
11 Mins for Hard Boiled

Kalien started working on the XPI installer and working through the XUL. We wanted an egg that showed up near the RSS tag on the status bar. ... it was hard work trying to debug those JAR files....

With Kalien working on the XUL/XPI stuff, David started working on Java script, Graem started working on making Egg PNGs, Rob was our Manager, and Jordi was the Catman.

And I.. well I was behind the Camera. actually later on I was working the firefox-egg logo, as well as trying to figure out XUL... but basically I was falling asleep due to the fact that I never really slept much this past week.

Once David completed the basic code for the timer, we decided we should try out our timers. Heh, Dave even made a buzzer. Although its not in Version .1, it will be someday! So on to making an egg....

Once the egg was done, we gave it a try.. the first Egg was a little 'underdone'... I guess we didn't submerge the egg in enough boiling water.

Our second try involved more water, as well as 'Hard Boiled' mode. 11 Mins later...

The shell came off good, now lets see how the egg tastes!!

Yummy!!!!

Now we just needed to make a good Firefox-Egg PNG and get the installer working... Few more hours pass by and w00t, the About Box shows up!

(Obviously as you can see, we have fallen asleep by now... 4:30am...)

Then making the Firefox with the Egg in Photoshop...

About this time it was nearly 5:00am. David and I were both passing out pretty quick, so we decided to Jet. On the way home I finished the Logo! When I woke up the next morning, I recieved an E-mail that said that Eggon .1 was released!!!

So Eggon! come and download it at http://www.rely.ca/eggon/