I have no technical backing for this suggestion except that I’ve seen it work on two different operating systems with the Google Chrome web browser.

Accessing my self-hosted WordPress blog to add posts and do site maintenance, page loads were taking FOREVER.  The media uploader was hanging too.  I thought it was a plugin hogging resources so I disabled a bunch of them and it still didn’t really help.

Finally I cleared the browser cache and cookies and now everything is back to normal speed.

Ideas anyone?  I’m running plugins only provided by wordpress.org and the latest  version of WordPress and everything else.

Mortgage Madness

I refinanced again–once again with no closing costs while sacrificing some on the rate.  Interest rates are again at historic lows which meant I had to take avantage of two great opportunities–lowering my rate by three quarters of a percentage and shortening the length of my loan to a 15 year term.  With no out of pocket costs involved (or closing costs financed), there was nothing to lose.

The upside is that I keep making the same payment I’ve always made and more money goes to driving the principle balance to zero because of the lower rate and shorter duration of the loan.  Two great sites I found for following rates and the mortgage process are:

• The Basis Point
• Mortgage News Daily

Collecting and Sending all those statements

Usually I rely on my tried and true fax machine, but this time I decided it was time to join the rest of the world and managed to collect everything as PDFs. After gathering them all in a folder I realized I didn’t want to send them unencrypted over the wire.

Next I searched for a way to zip all the files and lock the archive with a password that couldn’t easily be hacked and read on Windows. From what I could find on Google, the regular linux zip password protected algorithm was not secure, however encrypting with AES was.

I discovered that 7zip could do strong AES encryption and store in a zip format–presumably most Windows users (and my loan officer) would be able to read the archive, however the instructions I found to do so were horrible.

After google failed me I turned to the powerhouse of Red Hat engineers on an internal list… and in less than 24 hours I had a solid answer–thank you Norman Mark St. Laurent!  There is very little coherent information on Google about using 7zip with RHEL–the package names are confusing, the man pages are incomplete, and none of the examples I found for other platforms worked.  This post aims to remedy that.

Creating an AES encrypted archive with RHEL 6 and 7zip

1) Make sure EPEL (Extra Packages for Enterprise Linuxe is enabled)

2) Install the 7zip package (note the package name is not the same as the executable).

$ su -c 'yum install p7zip'

3) Create an encrypted archive readable by WinZip.

$ 7za a -y -tzip -pMY_PASSWORD -mem=AES256 archive-name.zip /path/to/directory-of-files

Image by Alexander Stielauvia flickr used under a Creative Commons license.

This quote from Mark Sigal’s Ruminations on the legacy of Steve Jobs was interesting,

If this seems intuitive, and it should be, consider the modus operandi that preceded it. Before Apple, the hard truth was that the “inmates ran the asylum,” in that products were typically designed by engineers to satisfy their own needs, as opposed to those of the actual consumers of the products.

Moreover, products were designed and marketed according to their “speeds and feeds,” checklists of attributes over well-chiseled, highly-crafted outcomes. And it didn’t really matter if at each step along the value chain the consumer was disrespected and disregarded.

Is this why “the year of the Linux desktop” hasn’t arrived or why after several year of running Linux, I’ve gravitated  towards Apple products (MacBook and iPad) for consumer use–outside of work?

To suggest that engineer led products haven’t turned out well is too broad of a generalization.  This quote could also possibly be misconstrued to say that notion of Open Source folks “scratching their own itch” is misguided and cannot create something great.  The Linux kernel itself refutes that notion, though it is interesting that Linux’s greatest success seems to be at the enterprise infrastructure level and not in the consumer mass-market.

Image by Alexander Stielauvia flickr used under a Creative Commons license.

Today I ran into a strange problem where messages seemed to be missing from my Inbox. Running “repair folder” from the properties tab automatically moved several messages to the trash. Moving them back to the inbox and repeating the process resulted in the same behavior. I thought maybe it was time to move on from the default version in RHEL 6, presently thunderbird-3.1.15-1.el6_x86_64, and live more on the edge.

I’ve also been using the same Thunderbird settings directory for two or three years so there was also the chance of some cruft build-up there as well.  What better way to rule out all possible issues than to run the latest version of Thunderbird with a fresh profile?

A simple way to do this is to rename ~/.thunderbird. If you change your mind or want it back, it’s simply a matter of closing Thunderbird, changing the name of the directory back, and running Thunderbird again. I believe there is also a way to this with the profile manager, but I like this way best.

I ran into one small problem due to my machine being 64bit and the regular Thunderbird Linux download being 32bit.  The solution was in this helpful forum post.

RhEL 6 Thunderbird Installation Steps

1) Download the latest stable Thunderbird version.

Updated October 27, 2011:  Thanks to the comment from Matej, pre-built 64bit packages are available, I’m not just not sure how regular folks like me would know to find them.  Back track on the link if you want a different language.

2) Un-tar the file that you download.  For simplicity I prefer to put it in my home directory.  With this approach it will not be accessible to other users, however it will not conflict with the existing installed RHEL rpm version.

$ cd Downloads
$ tar -xvf thunderbird-7.0.1.tar.bz2 -C /home/myuser

3) If you are running 64bit RHEL, you may encounter this error message:

thunderbird-bin: error while loading shared libraries: libdbus-glib-1.so.2: cannot open shared object file: No such file or directory

Installing the dbus-glib.i686 package provides the library you need:

$ su -c 'yum install dbus-glib.i686'

4) Run the new version of Thunderbird

$ cd /home/myuser/thunderbird
$ ./thunderbird

5) If you want to modify the launcher to run this version instead of the default installed RHEL version, right click on the launcher icon, select properties, and change the path to point at the Thunderbird binary in your home directory.  Something like: /home/myuser/thunderbird/thunderbird

Nostalgy is the most important add-on

The Nostalgy add-on is the primary reason I can’t bring myself to change email clients.  The ability to file messages with ease into any folder with one or two key strokes is something I cannot live without.  Don’t miss “shift + s” to save a message to the last folder you saved to.

Benefits of the latest version

• Once you set the column view you like you can apply it recursively to all your folders. This was a huge pain in the past where columns had to be set manually for each folder.
• Newer themes work and are compatible
• Everything feels polished, newer and faster

Use at your own risk

The downside of running Thunderbird this way is that you aren’t running the officially supported version provided by Red Hat and you are relying on Mozilla to provide the latest version to you.  This should happen automatically via Thunderbird’s built-in updating service.  Preferences for automatic updates can be changed in the “Advanced” tab.  Naturally the packages built and provided by Fedora and RHEL have this feature disabled.

I’m currently running the latest version of Firefox the same way and have not encountered any issues.

Image by Tonton Jaja used under a Creative Commons license.

My recent post on fixing terminator was months in the making, partly because I lost the environment I was using to double-check all the steps and partly because other things just kept coming up.  Several months back when I found the original fix it was only available as a patch in bugzilla.  I’d constructed a whole article about how to get the patch, update the spec file, build a new rpm, etc. but in the midst of this I lost the virtual machine I was using twice to a suspend that would not resume.  I also ran into some problems using yum-downloader to get the source rpm I could not reliably reproduce or document.

Time went by and an updated source rpm was available to fix vte so it seemed silly to document all the tedious steps to patch the spec file and rebuild when that work was already done, but then I was thwarted again when I rebooted my laptop and forgot I had a guest running.

It was a frustrating situation from the virt-manager GUI and the command line–my only options were resume (which didn’t work because of the error message below) and shut-down (which did the opposite of what I wanted the virtual machine to do).

On Google I found scant references to this situation so I turned to a company-wide mailing list at work where anyone can post technical questions.  I had the solution in 30 minutes!

This was the cryptic error message I received in virt-manager when trying to resume the guest:

Error restoring domain: Unable to read from monitor: Connection reset by peer

Traceback (most recent call last):
File "/usr/share/virt-manager/virtManager/asyncjob.py", line 44, in cb_wrapper
callback(asyncjob, *args, **kwargs)
File "/usr/share/virt-manager/virtManager/asyncjob.py", line 65, in tmpcb
callback(*args, **kwargs)
File "/usr/share/virt-manager/virtManager/domain.py", line 1050, in startup
self._backend.create()
File "/usr/lib64/python2.6/site-packages/libvirt.py", line 510, in create
if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self)
libvirtError: Unable to read from monitor: Connection reset by peer

The error message made no sense to me--I can’t resume the domain because the monitor cannot be read?  How about, “Unable to resume domain.  If this persists try ‘virsh mangedsave-remove’ from a command line to remove the suspended session and reboot your machine.” Including the word “suspend” in the command instead of “managedsave” would also be more intuitive.

Here is the command to remove a corrupted suspended virtual machine session so you can boot your machine again–naturally you’ll lose the suspended session you had:

$ su -c 'virsh managedsave-remove <domain name>'

I’ve also confirmed on RHEL 6.2 (beta) that rebooting the hypervisor with a running guests, smoothly suspends and resumes them again when the hypervisor returns.

In my switch from Fedora to Red Hat Enterprise Linux (RHEL) one of the things I lost was Terminator.  Even though it was in EPEL and installed without issue, the cursor wasn’t visible, making it hard to use.  For a while I put up with it.  After a while it drove me crazy so I fell back to Gnome Terminal.  And that was fine for a while until I started doing more command line things again and missed a working in Terminator so badly that I asked Google how to fix it.

It’s a bug in VTE

Wouldn’t you know, other people had the same problem as noted in bug 658774.   It turns out the problem is not with Terminator, but in the VTE package.

I’ve seen the problem on RHEL 6.0 and 6.1 and it looks to be fixed in RHEL 6.2.  It is presently available in the RHEL 6 Fastrack channel (if you have a RHEL subscription) as an update (login required to view).  I’d assume, but haven’t verified, that the same problem exists on CentOS 6.

If you are running CentOS and have the same problem or if you’d like to build a fixed package on RHEL yourself, here’s how.

Building a new VTE package with the fix

To build a new rpm with the fix you’ll need:

• a non-privileged rpm build environment
• the updated source rpm for VTE
• required build tools and libraries

Set up a non-privileged rpm build environment

If you don’t have a local rpm build environment specified, these commands will create it for you in your home directory.  There are several ways to do this, I found this somewhere long ago (maybe it was an RHCE class?) and like it because it is simple. You should always build the rpms as a non-privileged user–not as root.

$ mkdir -p ~/rpmbuild/{BUILD,RPMS,S{OURCE,PEC,RPM}S}
$ echo "%_topdir $HOME/rpmbuild" > ~/.rpmmacros

Get the latest vte source rpm

$ wget http://ftp.redhat.com/pub/redhat/linux/enterprise/6Workstation/en/os/SRPMS/vte-0.25.1-7.el6.src.rpm

Install the tools and libraries you need to build

$ su -c 'yum install rpm-build yum-utils gcc'
$ su -c 'yum-builddep ./vte-0.25.1-7.el6.src.rpm'

Build the new package

$ rpmbuild --rebuild ./vte-0.25.1-7.el6.src.rpm

Install your new vte rpm from ~/rpmbuild/RPMS

$ su -c 'yum localinstall /home/nopriv/rpmbuild/RPMS/x86_64/vte-0.25.1-7.el6.x86_64.rpm'

Run Terminator and you should be back in business.

It seems like the right time to make this move.  I am not involved in Fedora to the extent I used to be so I’m removing my full blog feed from Fedora Planet. I like all the traffic Fedora Planet provides and yet it doesn’t seem fair since most of my posts are not Fedora focused any more.

I will keep syndicating some content to Fedora Planet, mostly along the lines of anything related to “open source software”–broad enough to capture Red Hat Enterprise Linux, Fedora, open sources communities, etc.

If you want to continue to see everything I post be sure to subscribe to my main feed in your RSS reader.  If you’d like email updates instead, sign-up in the upper right hand corner at JohnPoelstra.com.

Image by Alexander Stielauvia flickr used under a Creative Commons license.

Riffing on a theme from Rework… I believe the project management structure and process framework of team should be a little less than “just enough.”  Anything more wastes people’s time and becomes more about the process than shipping (thank you Seth). Even though it was frustrating at times, Fedora taught me this lesson well.

There is little joy or satisfaction experienced on projects with oversight that amounts to executing the equivalent of the Space Shuttle’s pre-launch checklist on a toy airplane flight.

Image by keeping it real via flickr used under a Creative Commons license.

I’ve been a big fan of python bugzilla for a long time and used it often in Fedora to gather bugs for the release blocker meetings.  If you aren’t familiar with python bugzilla, it is a very useful library and command line utility that talks to the Fedora and Red Hat Bugzilla instance–a defect (bug) tracking system.

I tried to use it yesterday with Fedora 15, but it didn’t return any bugs.  James Laska pointed me at some unreleased patches and the commands to make it work again.  It doesn’t appear that any updates have been made to this package since August 2010, so hopefully a new release is on the way.

Here are the steps I followed:

$ git clone  git://git.fedorahosted.org/python-bugzilla.git python-bugzilla.git

$ cd python-bugzilla.git

$ curl "https://fedorahosted.org/python-bugzilla/attachment/ticket/32/0001-Fix-for-ticket-32.patch?format=raw" > /tmp/ticket32.patch

$ git am /tmp/ticket32.patch

Now, run the patched version from your git directory:

$ ./bin/bugzilla

A good example to try out is the example provided in the Fedora Blocker Meeting SOP.

I couldn’t be prouder of today’s OpenShift (PAAS–Platform as a Service) cloud announcement by Red Hat. It’s not often that you get to be project manager on a release this big or exciting. It was a massive team effort involving many smart and driven people at Red Hat and it was inspiring to work along side them.

If you’re looking for a free place to host your unmodified PHP, Ruby, or Python application, look no further than OpenShift Express.  See the OpenShift site for for more information about Express and the other offerings.

This is the culmination of one of the roles I assumed after transitioning from Fedora in November 2010.  Another project I took on will also soon release in Red Hat Enterprise Linux 6.1. It is an elegant, comprehensive certificate based system for managing your Red Hat subscriptions that goes beyond the existing Red Hat Network. This project was also particularly interesting and challenging since it involved a number of groups inside Red Hat I haven’t worked with before.

Speaking of Fedora, I hear they are on track for shipping Fedora 15 on May 24, 2011! I’ve enjoyed being obliviously unaware of the trials and travails of this release cycle while appreciating the very polished and stable Fedora 15 beta release. I wasn’t too sure I would like GNOME 3 after what I’d heard about the laptop power experience, but I do like it. I most miss the ability to see multiple time zones under the calendar and also find the calendar dim and hard to read at a glance.  For now they are minor inconveniences.