[digital:meditation]

Having been a merry Ubuntu GNU/Linux user for a couple of years now and, lately, kind of “flirting” with OpenSolaris for some reasons, these days once again I experienced what I like about technology, once in a while – the feeling of simply being stunned by the presence of a feature which might be obvious yet maybe not obvious enough to see wide-spread adoption: About to set up a server based upon the CentOS 5.3 GNU/Linux operating system (which, basically, is a “community rebuild” of RedHat Enterprise Linux), I fired up the network boot CD (dang, these old IBM xSeries machines didn’t yet come with a DVD drive…), chose to do the “text based installation” (as I usually dislike GUI based operating system installers and try avoiding them as good as somehow possible), made my way through configuring the base system, networking and mirror access… to, then, be asked whether I might want to continue installation using the “graphical installer” via VNC.

VNC? Well, why not… Accepted this, so it just took a couple of seconds until the installer provided me with the IP address and port of the VNC server running on this host, and, back to my notebook, connecting there I indeed was capable of performing the server installation completely from the comfortable environment of my office rather than standing in front of the machine in the loud and cold server room. So… I have to admit that this has somehow changed my mind about “GUI based installers”, and this definitely is a ‘killer feature’ from my point of view. Hope other distributions to come up with something like this (maybe an installer via ssh?) sooner or later, as well… ever even thought about doing a Windows Server System remote installation via RDP? Wonder whether this even would be possible… ;)

Some evidence screenshots, just because: Network and disk configuration using CentOS VNC installer. Pretty neat. :)

Faster than I thought… Fixed my broken X server pretty quickly, and by the way learning how to boot up OpenSolaris into maintaineance / single-user mode (as this unfortunately is not there as a default recovery option in the grub configuration but rather easy to do nevertheless). Using the xorg.conf I copied off my Xubuntu system, now my external TFT finally works at 1280×1024 / 75 Hz, took some time to get that figured out in GNU/Linux as well. Adding a custom modeline to xorg.conf was the way to go, and this worked flawlessly on OpenSolaris’ Xorg as well, same as using xrandr to switch to the desired mode. Good.

Along with this, figured out that the Eclipse IDE in its latest (3.5) release finally is available to Solaris/x86 users again, which is pretty good – even though Eclipse is not my favorite tool getting development work done, I have little choice regarding this in a current project so having this tool available on OpenSolaris makes the platform a more likely working environment again. Downloading, unpacking and starting worked same as straightforward as on all other platforms, no problem as well.

Also managed to, again, get my fonts set up to be equally sized in Firefox/Thunderbird and the rest of the UI – setting a different resolution (obviously it’s 100 not 96 dpi) and playing around with the font anti-aliasing options in the GNOME theme configuration tool did have the desired effect. Good to also have this resolved, even though it just was a minor annoyance, and even though fonts on some web pages in Firefox still look kinda borked. :)

Installing OpenOffice, NetBeans and friends off the IPS repository by now, although once again I quickly dumped the package manager UI (which, even while attempting to get a first package installed, locked up and refused to refresh pretty quickly) in favour of the pkg command line tool which seems to work much more flawlessly. Let’s wait and see… :) Minor annoyance detected here: The Gimp, another application I (for obvious reasons. ;) ) excessively use, is just to be found in the repositories in a rather old (2.4.6) version. Maybe this is subject to change in the current development builds…

Following an extensive discussion on osnews.com evolving around the question whether Suns OpenSolaris distribution is just a bad GNU/Linux distribution, I once again did a chance downloading a recent build off genunix.org, installed with the motivation of trying to use it in everyday work for a couple of times to see where it gets me. As I usually do my everyday work exclusively using xubuntu, all comparisons just will relate to features that this platform provides, knowing that I use this distribution because at the moment it fits all my needs so there, generally, is nothing “more” I need at the moment, leaving these features being met the minimum requirements talking about OpenSolaris.

So, starting this, a few first impressions after successfully installing this ISO:

• Installation speed seems to have somewhat improved; overally, on my machine this time it was done in under an hour (and I once again had to burn it to a DVD medium because the ISO didn’t work copied to an USB stick using unetbootin.
• Even during running the live system and doing the installation, I merrily figured out that by now I am finally able to access my WEP encrypted home WLAN using the OpenSolaris network manager (which, in earlier versions, failed for whichever reasons). So one thing being a big issue to me in the past seems gone. Good work.
• Starting up the system for the first time, I again had the comfortable feeling that speed has improved, both talking about bootup and logging in to the GNOME desktop which is the default (and only?) desktop environment supported out of the box. Speed also was an issue in the past at times, so this is generally good, although I have to see what happens if the system is “under load”.
• This morning, however, I had to figure out that OpenSolaris, though providing a display configuration dialog, doesn’t out of the box allow for using a desktop that spans two different displays – although both my external TFT and my notebook LVDS display were correctly recognized in terms of resolution and frequency, I figured out that after setting things the way I wanted them to be, the tool eventually complained that the maximum allowed screen size is smaller than the extended screen I was trying to create. Oh well.
• A minor issue that still is unresolved in my opinion is that, talking about desktop applications, fonts look slightly and in a subtle way different in applications like Firefox or Thunderbird and the rest of the screen. This is strange and I remember fighting with this in the past, so I will have to remember how I resolved this earlier. At startup, it doesn’t really look homogenous.
• The only issue left during installation that really is annoying, always was and always will be, is that OpenSolaris, same as blind as another operating system, quietly discards any currently installed bootloader and replaces it with its own grub installation, pointing to the OpenSolaris installation and to some imaginary “Windows” installation that doesn’t at all refer to a meaningful drive or partition. I can’t really imagine what could be so hard about making the installer at least ask whether or not to install a boot loader and/or letting it include existing operating systems correctly.

So much as a start, seeing that some has changed for the better and some hasn’t. More to come later, as soon as I managed to restore my X installment which I broke somewhere all along following this blog post regarding multi-monitor setup. Shouldn’t be too hard, then again. :)

So once again, another Java User Group Saxony meeting is over: Yesterday, on July 16th, our seventh evening of lectures and communication took place, once agan located in an auditorium of TU Dresden Faculty Of Computer Science, and roughly 30 people (less than in earlier meetings, but still a good number considering the time of year and the overally warm weather…) came to listen to two pretty high-quality lectures announced to be focused on the Groovy programming language and its use in real-world applications. In the end, it turned out to be a little different than that but very interesting nevertheless…

(more…)

Fought a half-day fight battling with our local Cyrus IMAPd installment. And talking complexity again: We’re using right this mail server implementation because it provides extensive support for shared folders and per-folder access control lists allowing for pretty fine-grained, well, control of user access to shared mail folders. ACLs indeed do add complexity to that, however, and this is what, in the end, seems to have enabled one of our internal user to move part of our IMAP folder tree to her personal Trash folder. Gone?

Fortunately not, given the Mozilla Thunderbird client involved here did just move things there and mark them as ‘deleted’ but not actually expunge them. Copying things manually out using a shell or some other file manager was pretty easy, and figuring out what (cyr)reconstruct is supposed to do also wasn’t a job that though… to overally end up with most of the folders restored but most of them empty?!

Googling around and looking deeper into things, however, quickly did resolve this issue as well: Cyrus uses to store all its stuff in /var/spool/cyrus/mail/, sorted in a more or less straightforward hierarchy. And this is where my problem finally resolved:

• Mailbox Storage lives in /var/spool/cyrus/mail/s/Storage. So far, so good.
• Mailbox Storage.Contacts, however, lives in /var/spool/cyrus/mail/c/Storage/Contacts… which I missed paying attention to, in this situation.
• After having the mess moved to Trash manually, all the hierarchy was stored in /var/spool/cyrus/i/user/involvedUser/Trash/, including all subfolders. So just copying it back to its most obvious place (.../mail/s/Storage) didn’t work out as expected…

So after another examine-and-move session (copying all the folders starting with “A” from the deleted storage hierarchy to /var/spool/cyrus/mail/a/Storage/, all the “B” folders to /var/spool/cyrus/mail/b/Storage/ and so forth and finally running a cyrreconstruct -rfx Storage.*, both the hierarchy and the mail folder content seem to be back. And I am pondering a more sane mechanism of backing these things up… at the moment, without a good file manager I surely would have been doomed, so looking back at this match: Cyrus 0, gnu mc 1. Been there, done that. :)

As some might know, I have been involved with the NetBeans Community Docs project for quite a while now. I have been using NetBeans IDE same way for quite a while, and I used to contribute a couple of documentation and articles on “interesting” topics to the NetBeans Community Docs project in the past. By now, however, I can’t help feeling the project slowly loosing momentum, I have a couple of vague ideas why it is this way, and even less ideas what to do about that…

(more…)

Starting with this blog post regarding the inclusion of tomboy note-keeping application into the default installation of GNOME desktop on Debian GNU/Linux, quite a discussion has started whether or not free software should depend upon Mono and/or mono should be included in a default Debian installation. Mono Project, as some might know, is aiming at being an open-source re-implementation of .NET framework and technologies originally created and provided by Microsoft, partly being covered by more-or-less “open” standards like ECMA C# and CLI, partly being suspected of being protected by software patents thus eventually being difficult used along with free-as-in-free-speech software.

Generally, I don’t really want to go into bothering about the technical and legal aspects of this discussion, as enough has been written on that so far already, and I don’t really know what else to add to this… except for one thing: In some way, this whole mess is rather irritating to me, as it seems to show that, in this field of technology, Microsoft has won after all… It’s not about .NET (not) being a good technology – actually, I think a lot of ideas provided by .NET and/or Mono are rather good even looking at it from a Java developers point of view. What really bothers me here: Making a “software libre” desktop system built upon the re-implementation of Microsoft technology feels utterly strange. It feels like, while most of “the software development world” has moved on the last couple of years providing new concepts and technologies, the “software libre” folks thoroughly have missed creating a development environment more sophisticated and reliable than old-fashioned C/C++ for most of the doings. So what remains in the end? No matter why you want to choose an open source or even “software-libre” desktop, maybe even for the reason of not depending upon Microsoft technology (for whichever reason you don’t want that), you will suddenly see that a whole load of applications (like tomboy, f-spot, banshee and a growing bunch of others) in this “supposed-to-be-free-of-Microsoft” world actually are built upon Mono which at its very heart is based upon a Microsoft invention. This eventually is not the message “software libre” distributions want to send out to the world, even though .NET eventually is the best thing Microsoft have come up with so far…

As I am usually doing day-to-day work using daily builds of the NetBeans IDE, then and now I happen to run into, let’s say, peculiar situations in which figuring out what actually happens is pretty difficult. My very special friend #162706 is one of these cases – the IDE just seems busy for quite a while without obvious reasons, and you’re not really sure whether there still is something goin’ on worth waiting for it to come to an end…

After more or less loosely following Geertjans blog I rather early found out about the jvisualvm tool that comes with recent versions of Sun JDK 6, but so far I haven’t really seen any use cases for it in my environment. However, at the very least now I have figured out that this is rather good a solution for tracking down weird NetBeans behaviour by providing more reliable information, i.e. thread dumps. And there’s even more…

(more…)