Technology Tales

The peril of /tmp

By default, I think that Windows plants its temporary files in c:\windows\temp. On Linux or on Ubuntu at least, the equivalent area is /tmp. However, not realising that /tmp when you shut down and start your PC could cause the silly blunder that I made today. I was doing a spot of reorganisation on my spare PC when I dumped some files in /tmp from a hard drive that I had added. I was reformatting the drive as ext3 following its NTFS former life. As part of this, I was editing fstab to automatically mount the thing and a system restart ensued. I ended up losing whatever I put into /tmp, a very silly blunder. Luckily, I had the good sense not to put anything critical in there, so nothing of consequence has been lost. Nevertheless, a lesson has been learnt: Windows allows its temporary area to pick up all kinds of clutter until you clear it, while Linux clears the thing regularly. It's remarkable how Windows thinking can cause a howler when you have a lapse of concentration using a *NIX operating system, even for someone who uses the latter every day.

Technical considerations regarding the discussion aspect of blogging

When making a start in the world of blogging, there are so many things to consider that you almost need a trial run first to learn the lingo. In fact, getting up to speed by using a service like that offered by Blogger or WordPress.com seems a very sensible starting point. Even so, the business is like building a house in that you only really know what you are doing after you have done the deed and made all the mistakes. That is particularly true when you go down the self-administered blog route. For starters, it's so easy to pick the wrong domain name or hosting provider. Selecting your blogging software is the next step, but that may not be so tricky; WordPress does a reasonable job and there's always Movable Type, Expression Engine, Drupal (yes, really) or Habari.

That mention of blogging software brings me to something that I encountered recently: commenting functionality. I am coming around to the idea that this is probably something that needs to be considered up front because of the nature of blogging. The Blog Herald's regular readers should understand blogging conversations; consequently, the technology must be easy for visitors to use and simple for bloggers to administer. However, the two can collide. For one thing, there are a myriad of choices available to the blogger, while the blight of comment spam is ever pervasive and growing.

When it comes to comment spam, it is best to realise that there are two sources of responses to a blog post: visitor comments or trackbacks (pingbacks?) from other blogs. As it happens, I reckon that the latter is probably the channel where most of the detritus travels, with various anti-spam solutions are on offer to curb its spread. Names from the WordPress world like Akismet, Spam Karma, Simple TrackBack Validation and Bad Behaviour come to mind. The former can also be used, particularly when the unscrupulous make use of low-cost labour in low-cost countries, and that's when the thorny questions of user registration and CAPTCHA's arise. There is something to be said for not going to extremes with these and just sticking with less onerous rules and filtering on the server side.

Admittedly, I have stumbled into forcing visitors to register before adding a comment and then making them log in thereafter. While I think that it's for security reasons, WordPress creates a password and then sends it to the person who is registering rather than displaying it on a web page. That can create another problem: what happens if the email fails to arrive? In the last week, this has happened with a visitor to my hillwalking blog.

There could be a number of reasons for the non-arrival of the relevant email. One is ironic: being an automated email, it gets stuck in the spam filters of the recipient's mailbox and so never gets to them. It could also be a bug with WordPress itself (I have raised a ticket, and I am awaiting what Automattic might have to say to it) or a consequence of some setting made by a hosting provider. All of that makes it challenging to track down the cause of the issue, yet it kicks off other thoughts as to its resolution. One is to remove the needed for registration and logging in the first place, but there are third-party services that may help too; the former has turned out to be the case for this place. Since it seems to be performing well enough, it is an acceptable option.

When it comes to using third-party comment handling systems, what needs to be considered is how well they work with your blog. For instance, I gave Disqus a quick whirl and soon realised that I needed to update the themes for my WordPress blogs if I were to use it on an ongoing basis. While it worked fine, I was left wondering if it would have been better to have brought it in when I started a blog rather than part way through and with comments made using the existing WordPress functionality. Intense Debate is another option, one of many if my hunch is right. The theme for my hillwalking blog has been modified to allow prospective commenters to get in touch with me if they are having problems. That is only an interim approach while I consider what the way forward will be.

New version of my Countrytones plugin

Now that WordPress 2.6 is out, it is time to introduce a new version of Countrytones to the public. A few CSS tweaks have been needed to the original version after the changes that have been made to the administration interface for 2.6. Those screens still look largely the same with this release as they did before, but for the styling of things like the bubble that alerts you to the availability of plugin upgrades, among other things.

Download Countrytones 1.0.1

Disk corruption can be virtual too

It's the sort of sight that causes you to fear the worst, an unchanging black screen with a flashing cursor. That was what started to greet me recently when I tried to fire up a Windows XP guest in VMware Workstation. The cause was the corruption of a virtual disk, an ominous thing that affected a supplemental virtual hard drive that I had added to the virtual machine rather than its main one. Though there might have been some data on there, it was nothing that I didn't have elsewhere anyway. Removing the broken disk from the XP VM returned the situation to fuller health, and I simply tried creating a new one again. So far, this seems to be working fine, yet I'll be keeping a watchful eye on things.

So far, I have no idea why the corruption happened, but the broken disk files were created without a VMDK extension, which might indicate that something was amiss with the process that created them. While it would be better if VMware highlighted the state of its virtual disks, it was when I tried opening the trashed disk with VirtualBox that a warning was given and VMware did the same when I tried it with that afterwards (opening VMware virtual disks with VirtualBox is possible, but you need to ensure that no attempt at importation is made, or you could end up with a breakage). While I may have discovered the fault in a roundabout manner, it's better to know what's gone wrong than not to know at all.

WordPress 2.6 out in the wild

Though there was a time when I tested out a new WordPress release when it made its appearance, I now keep tabs on development versions too. It helps me to be ready in ample time and avoid any ugly stuff appearing on either here or my hillwalking blog. On the surface, the changes made don't seem so dramatic after the revamp of the administration interface that came with 2.5. Under the bonnet, many bugs have been fixed, and many behind-the-scenes features have been included for the first time. I'll leave it for you to go to Planet WordPress or check out the relevant entry on the WordPress Development blog. You'll find a Flash video tour of the latest version from Automattic below.

Of course, WordPress development doesn't stop here and there are some considerable changes to the administration interface to come at some point if they ever make out into the mainstream from the crazyhorse branch. I'll be continuing to keep an eye on how things go from here to be ready for whatever might be in the offing. For now, enjoy 2.6, though it'll be interesting to see where it all goes from here.

An introduction to Wubi

The Toshiba laptop that I acquired at the start of the year is a Windows Vista box, and it isn't something with which I want to play too roughly because the OS came pre-installed on it. I still wish to continue to see how Vista goes at close quarters, so removing it to put Ubuntu or some other Linux distribution on there wasn't ever going to be an option that I was willing to take either. Neither was the option of setting up a dual booting arrangement using disk partitioning; I have plenty of experience of doing that to set up dual booting machines over the years, and I don't need any more than what I already have. So, I was happy to leave it as a Windows box and only as a Windows box.

That situation has changed, and the cause was Canonical's decision to go for something novel when it brought out Ubuntu 8.04. The premise is as follows: a Windows style installation that pops an entry in the Windows boot menu that allows you to fire up Ubuntu without ever having to do disk partitioning or other similar rough play. For those who are less than enamoured with the Linux option, it's even easy to remove too, as easy as any other Windows program in fact. Removal of Linux is very definitely not what I'd do, and that's even without the pain and upheaval of more customary means for setting dual booting machines. In these days of virtualisation and hypervisor technology, I have my ideas as to what has been used to give us that easy way in.

Being an Ubuntu user anyway, the possibility of having Ubuntu on the laptop and the interesting opportunity that Wubi offered for getting it on there was too tempting for me to give it a miss. A small download from the Wubi website is all that is needed to set things off. You get a number of options up front, like where to put the (large) file to be used to house the Ubuntu world and how large you might want it. Setting a username and password for the thing gets included among other items. The next stage is to download the files to be used to perform the installation. Once that is completed, and it took me a few goes to get the lot (thankfully, it stores things up to the point where the downloading operation cuts out so you didn't start from scratch each time; even so, it's still annoying and could put some off), it is time to restart the computer and boot into Ubuntu to complete the set-up of the operating system itself; it is at this point that the familiar very much returns. A reboot later, and you are into a world that does its level best to fool you into thinking that Windows is another universe and never existed on that machine at all.

So, a machine that seemed destined to only ever run Windows can run Linux now as well. Wubi comes across as a neat and clever way to get a dual booting computer, and I hope to leave mine as I now have it. No feathers were ruffled on the Windows side, and I saw no sign of any destruction. That makes Ubuntu's way of doing things a much better option than other distributions that make you go down more invasive routes when creating a dual booting PC. A question remains in my mind. Could this approach take off?

Watch where you store your virtual machines when using VMware on Linux

My experience is with Ubuntu on this one, but I have found that you need to be careful as regards the file system used by the drive where you keep your virtual machines. If it is NTFS, VMware can fail to start a VM because it cannot create a virtual memory file while it presents as physical memory to a guest operating system. Use ext2 or ext3 and there should be no problem, even if that means formatting a drive to fulfil the need. That's what I did and all was well thereafter.

A collection of lessons learned from using Eclipse on Ubuntu

I have been running into a few woes on the home computing front that may or may not give rise to a number of posts on here. While having my Windows VM's going awry on VMware is a more worrying development with my need to use a Windows-based application for my hillwalking mapping, I am going to devote this entry to a spot of bother that I started to have with Eclipse, if only because I managed to sort that one out.

Up to yesterday, I had all my offline website development stuff in a single project area for the sake of ease of testing. I suppose that I got led into this by my use of Dreamweaver and the way that it sets up what it calls Sites. Applying that same of working to Quanta Plus and NetBeans just chokes up the respective IDE's and makes them less usable. Until recently, Eclipse escaped this because it seemed to check if a file had changed when you tried editing it and asked you if you wanted the latest version. This stopped in the last few days for whatever reason, and it started to stall just like the others.

Naturally, I wanted to set it back as before, so a certain amount of investigation was in order. I ended up refreshing my installation in /usr/lib, a manual extraction of the Eclipse PDT archive, only for that not to resolve the issue. In fact, it created another one that we'll talk about later. Creating a smaller project made it all work again, and I'll be building up a number of these.

The new issue pertained to the creation and selection of the Eclipse workspace. While there was no problem using what I wanted it to use, it wouldn't remember the setting. There was more blundering about before I happened on the cause: access permissions. Then, I copied the new Eclipse files in as the root user, and that meant that Eclipse couldn't update its setting when I was running it under my own account. Running the editor using sudo sorted out the workspace selection issue for now, but a more permanent fixing such as giving myself write access to the configuration directory and its contents remains an outstanding task.

The mention of the Eclipse workspace brings me back to the way that it was working before the upheaval hit me. It does keep a copy of every file that you edit in there, and maybe more besides. Thus, having a copy of every file in the project would have meant that it didn't need to do the constant churning being performed by Quanta or NetBeans. That's the impression that I have, so I'll stick with smaller project bundles from now on. Learning all this was useful.

Running Internet Explorer on Linux

On first sight, this probably sounds daft given how good Firefox is, yet you cannot ignore those surfing the web using the ever pervasive Internet Explorer when doing some web development. Though using virtualisation is a solution to the need, it can mean that you need to set up a web server with Perl, PHP, MySQL and the like in a virtual machine, all for a little offline testing and then there's the potential for a lot of file copying too. Otherwise, you are trying to sneak things online and catch the glitches before anyone else does, never a good plan.

Therefore, having the ability to run IE to test your offline LAMP set up is a boon, and IES4Linux allows you to do what's really needed. Naturally, WINE is involved, so some flakiness may be experienced, even after the ever useful API library's reaching version 1. Otherwise, all usually runs well once you work your way through the very helpful instructions on the IES4Linux website. I did get a misplaced message about the version of WINE that I was using, and Python errors made a worrying appearance, but neither compromised the end result: a working IE6 installation on my main Ubuntu box.

IE5 and IE5.5 are also on offer if you're interested but, after looking at my visitor statistics, I think that I can discount these. IE7 and the work-in-progress IE8 make no appearance on the availability list. The absence of IE7 is not a big problem as it might appear because coding for IE6 sufficiently suffices for IE7, even now; IE8 may not be the same in this regard, but we shall see. Even so, a later browser release does mean a more secure version, and I reckon that including IE7 should be next on the project's to-do list. Saying that, what we have now is far better than nothing at all.

Killing those runaway processes that refuse to die

I must admit that there have been times when I logged off from my main Ubuntu box at home to dispatch a runaway process that I couldn't kill, and then log back in again. The standard signal being sent to the process by the very useful kill command just wasn't sending the nefarious CPU-eating nuisance the right kind of signal. Thankfully, there is a way to control the signal being sent and there is one that does what's needed:

kill -9 [ID of nuisance process]

For Linux users, there appears to be another option for terminating a process that doesn't need the ps and grep command combination: it's killall. Generally, killall terminates all processes and its own has no immunity to its quest. Hence, it's an administrator only tool with a very definite and perhaps rarely required use. The Linux variant is more useful because it also will terminate all instances of a named process at a stroke and has the same signal control as the kill command. It is used as follows:

killall -9 nuisanceprocess

I'll certainly be continuing to use both of the above; it appears that Wine needs termination like this at times and VMware Workstation lapsed into the same sort of antisocial behaviour while running a VM running a development version of Ubuntu's Intrepid Ibex (or 8.10 if you prefer). Anything that keeps you from constantly needing to restart Linux sessions on your PC has to be good.