2010-05-13 17:15:08 PDT - T[0x00007FFF70D74BE0] - DNSServiceProcessResult returned -65563

Like the rest of them, I’ve been wondering what this means and if it could be the cause of some of the problems I’m having. But now I don’t think it is.

After much searching, I finally found Apple’s API documentation for DNS Error Codes. Here’s an excerpt.

kDNSServiceErr_ServiceNotRunning = -65563, /* Background daemon not running */

Like me you’re probably pretty sure your DNS service is actually running. Now that I look at the logs, this only happens on startup. I have one server that’s running Open Directory and DNS so I can take a guess that Open Directory starts trying to use DNS before DNS has fully started. I’m only seeing this log entry when I restart my server.

I doubt this is the cause of any serious problems. I hope it’s not because the only way around it would be to buy another server to run DNS.

Here’s my situation, and I’m sure there are many other people that are in the same position as I am. One of my clients is a small to medium sized business and they use Macs. After I started working for them, I think they started to see I could help them with a lot of business tasks. After talking with me about their goals they decided to buy a couple of Mac Mini Servers.

I’m excited at the chance to set these up for them and I even bought one for myself which is already working out to be quite a time saver (well, not exactly at the moment, but it will be.) I’m a smart guy, I worked as a developer at Microsoft for many years. I knew I could do this server thing and it’s been a tough learning curve, but it will be fine. My main problem is that I’m having a hard time figuring out what an OS X Server network is “supposed” to look like.

With a Windows Server network, at least the one we had at Microsoft, you log in to each computer as your domain user, it verifies your password via the network, sets up a local home directory on the computer and you’re done. Whether the network is there or not, you can log in. Windows Server isn’t big on network home drives. People have tons of storage on their workstation, it’s too expensive to ignore that.

OS X Server is just the opposite, and it doesn’t make a ton of sense to me. They’re favorite thing is the network user with a network home. This is great because you can log in anywhere and your files are available to you. The problem is, what about the terabyte of storage on the local workstation that the user uses everyday? Apple doesn’t have any suggestions here, but I guess you have to educate the user on how to store things locally, because they aren’t going to have that much storage space on the network (for small businesses).

I found out how to have a home folder created locally on each computer a user logs into, which is kinda good. But what I would really like is some sort of integrated solution where a user has both network and local stuff and the idea of that is fully supported.

Also, the computer you logged into doesn’t remember you or your password if you have a network account. It might store some files for you, but if the network goes down, nobody in the whole organization can log in. For a small business with a part time administrator, this is not an option.

Of course, you can choose to have the users have a true local machine account (not managed by server) and a network account (managed by the server) and then log into services while logged into their local account, but this isn’t really supported either. These are truly two separate accounts and nothing is going to confuse users more than having two accounts and two passwords, especially if you do as some suggest and give the accounts the same username, which is totally possible. My clients would freak knowing that one username had two passwords that had to be changed separately in separate places.

So now I’m looking into creating mobile accounts. My understanding is that then the computer will remember you so you can still log in if the network is down. I can minimize syncing to certain folders and train the users that these folders are “network” folders. I can also train them to use their local drives for the files they only need on a single computer.

I’m so curious as to how Apple has their network set up. Does every user have two logins and passwords? Do they all have network accounts but aren’t able to log in if the network goes down? Do they all have mobile accounts for the reasons that I think they are a good idea? How do they make the most of both network and local storage?

The thing that’s most disappointing about Mac OS X Server is that it seems like this issue is the elephant in the middle of the room that nobody’s talking about. But it really needs to be talked about? Real businesses don’t just have library terminals set up everywhere. So how is it supposed to look?

If you have any knowledge that can help me find my way, please share it in a comment! Or if you just want to add to my rant, please feel free to do so!

Over the last couple of weeks, I’ve been taking a crash course in Mac OS X Server administration. I bought a Mac Mini Server and ever since I’ve been reading as much of the server documentation as I could and installing and reinstalling and staying up way to late every day.

I reinstall now about once every two days, and each time, I try to maintain as much of the previous installation’s settings as possible. For instance, last time I exported and reimported DNS and Open Directory (via Workgroup Manager) information successfully.

I keep network home directories, time machine backups, and mysql data on the second drive in the Mac Mini Server so that I can do clean installs without destroying that data.

As I’m geting more used to using the tools such as Server Admin and Workgroup Manager, I’m using Server Preferences less and less. This happened organically. Server Preferences has a simpler friendlier interface, but I’m finding that it may be a “false friend” as it sometimes does more than you’d like when you want to have finer control over your server.

After my most recent clean install, I enabled the AFP service in Server Admin and I was removing the default shares and adding my preferred shares on the other drive. I also went ahead and shared the network Time Machine disk. But when I started the AFP service, none of my computers that had previously been using Time Machine on the network acknowledged that the drive was there.

I went into my iMac and I re-selected the drive. It started a fresh backup but in the root directory of the drive instead of in Shared > Backups where it was previously backing up. I immediately stopped that backup and went to investigate.

For all previous installations I had enabled network Time Machine in Server Preferences, not Server Admin. So I went to Server Preferences and then to Time Machine and it told me something strange. It said that “File Sharing” was not turned on. But it clearly was. My network home directory was working fine and I had already started and stopped a backup using network Time Machine.

Well that’s it, clearly File Sharing is a different concept from AFP. First of all, File Sharing will share things by default using SMB and AFP, which at least at this point, I don’t care that much about. I can always enable SMB when I need it.

Also, when you start network Time Machine using Server Preferences it creates the directory structure Shared > Backups on the drive you select. Then it only shares the Backups folder. So when you go in Server Admin to do the same thing, you only want to share the Backups folder and not the whole drive. Once I did this my remaining computers woke up, recognized that the Time Machine drive was now present, and continued to back up as if I had never reinstalled OS X Server. This is just what I wanted.

Also, for the computer where I had changed the backup drive, I now selected the new share and it also recognized its old backup file and continued with its backups as if I had not reinstalled. This was a pleasant surprise.

So now my server is set so that “File Sharing” is off and AFP is on, and that’s fine with me. Hopefully I can forget that Server Preferences exists and it will be one less tool to have to worry about.

The operation can’t be completed because the item “filename” is in use.

I got it over a period of days even after switching files in the only application that could have possibly used it and quitting that application. When I looked for help on this, I saw only the obvious suggestions, “make sure you shut down apps that use the file”, to the crazy, “you need to repair permissions”.

Repairing permissions is like control-alt-delete for Mac users, except that it’s usually not necessary and rarely works but it’s the first thing Mac users from a certain age suggest, along with reboot. But in this day and age, I shouldn’t have to reboot to delete a file or do any other crazy stunts. Then the obvious hit me: use Terminal.

So I opened Terminal and typed this:

rm -f filename

It worked great. The rm means remove and the -f means force if you’re new to this.

You can drag files from the Finder into Terminal and it will insert the filename into Terminal for you. I used this trick because it was a long network path.

continued after image…

I’ve realized that MediaWiki is what I’ve always wanted from a specification writing tool. The limited but fast and flexible formatting options of a wiki allows you to really focus on your content. I think it’s also very handy that you can edit one section at a time. The wiki language is the fastest way to created tables that I’ve seen. I copied the common.css page over from Wikipedia so the tables look great because I add the “wikitable” class to them.

When I was at Microsoft we tried using some Windows specific wiki software for specifications at one time. It kind of worked, but it kind of didn’t. I think maybe just because not everyone had laptops at the time, so we had to do a lot of printing for meetings. It may have also been that the Microsoft wiki software sucked relative to MediaWiki. But having had this experience, I can’t imaging writing specifications any other way.

It’s great for more than specifications, like just keeping track of even unorganized thoughts that may be useful. Sometimes I will just figure out some random idea and need to write it down, but everything gets lost on paper, or even text editor files. As soon as you’ve opened up a word processor application just forget it, the moment is lost.

I think MediaWiki is my personal productivity application of the year. I have really never been this productive before.

Installing MediaWiki is theoretically and practically easy, but the first time still takes a few hours. I spent a good amount of time figuring out how to get pretty URLs which was complicated by the fact that my wiki was going to be at the local address http://wiki. Hint: Don’t name your internal wiki site wiki if you want the process of getting pretty URLs to go quickly. I renamed my wiki directory and site name to something else and proceeded successfully. Luckily, all of that while taking some time, went smoothly.

I have my wiki installed on an Apache virtual host, which the documentation writers don’t really think of as a common scenario. If you are going to do this and want pretty URLs, just realize you will have to know a little bit about .htaccess files. If it helps, here’s my .htaccess file I created as part of the process to get pretty URLs:

RewriteEngine On
RewriteBase /
RewriteCond %{REQUEST_FILENAME} !-f
RewriteCond %{REQUEST_FILENAME} !-d
RewriteRule ^wiki/(.*)$ /index.php?title=$1 [PT,L,QSA]

Overall I am happy with it, although in some ways I’m just mildly unhappy with it. I would prefer a web based administration page to manually editing the LocalSettings.php file. I would prefer that MediaWiki automatically create the appropriate .htaccess file for pretty URLs based on your installation. (Why would anyone not want pretty URLs?) On the other hand, the installation is pretty smooth and MediaWiki really keeps you focused on content because there isn’t much else to do but create content. This is a good thing.

I’ve said before that I used to work on Internet Explorer. But when I worked on it IE really was effectively the only browser out there. We actually took compatibility with Netscape very seriously, but then we went way ahead and implemented more features from the standards. The problem is, we made mistakes in those implementations and since there were no other browsers, our mistakes actually became a sort of standard.

But those days are long long gone. Now it’s far more often the case that my code works “the first time” on every browser out there except for Internet Explorer. The other browsers are all more standards compliant by far. They also score better on the Acid Tests, and that’s why I love the Acid Tests.

When asked, Microsoft will say that the Acid Tests are random, that they aren’t developed with real world scenarios in mind. There is a great Microspeak term invented at Microsoft: the Acid Tests are a forcing function.

A forcing function is a situation or constraint that forces you to make a decision you would not make otherwise. For instance, in my end days at Microsoft, while working on the Windows Presentation Framework, one of the high level managers told me to come up with a plan to implement a feature that I was very outspoken about not implementing. I was given a week (I think) before presenting this plan to a group of all the highest level managers in our group. They were very sneaky saying that they didn’t necessarily want to implement it but if we did, how would we do it? The idea was to get me to do the work under false pretenses, then they would force me to put it in the product since it was already done. While talking about this later, the term forcing function came up, I hadn’t heard it before that. That situation was one of the nails on the coffin of my relationship with Microsoft.

Now that I’ve finished with my Microsoft story, I’ll say that learning the term was the best part of the experience because it describes that kind of situation very concisely.

The snakes were the forcing function that made me decide to mow my lawn on a regular basis.

If you decide to go live on the 10th, that’s going to be a forcing function for about five other required sub-projects and your bill is going to be a little higher than usual.

So Microsoft, it’s time to take your own medicine. We don’t care if you don’t think the Acid Tests represent real world scenarios. The truth is, fixing these bugs to pass these tests will actually and truthfully make your browser better. All the other browsers have proven it’s quite possible. You’re not protesting, as I was, out of a belief that the forcing function will make you do the wrong thing. You’re protesting because you’re incapable of actually fixing your browser.

Microsoft, please get out of the browser business and make everyone’s life easier. I’m tired of waiting for you to figure out how to write software.

I’m a big fan of Google hosted email. You get that great Gmail experience, but with your own domain. It also incorporates Google Docs and Calendar right into the mix so companies can easily share documents and meetings. This is great.

But what happens if your company wants to use AdSense? Well, you have to “sign up” for a Google Account using your email. Then it will create an account for you. But wait, didn’t you already have an account? Yes. So now you have two accounts with the same email, and they are both google accounts, but they are not the same. The AdSense account will also hook you up with Google Docs, but its a whole different Docs than the other account. Confusing, huh?

Now add Google Analytics, Google Webmaster Tools, Google Voice, and Google Reader. They can only use one of the accounts as well.

Furthermore, lets say you had a thriving business and were actually making good money by participating in AdSense? Let’s say this account was under some Gmail address. Well, now you have this wonderful domain address, and you don’t really need that Gmail address. You want to transfer everything over. Well, too bad. You can’t. You just can’t. It’s not like you have to know the right people or something. Google just gets fixated on the email associated with these services and has no concept of ever changing that. They actually “recommend” you shut down the old AdSense account completely and start a new one. Oh, and by the way, the policy clearly states that you can’t have two AdSense accounts at the same time, so that should be a fun (and expensive) transition for you if you choose to make it.

This might seem odd to you. You might even try to search (using Google) for an explanation. But there isn’t one. Google is bizarrely silent on the issue. Well, you can imagine that one day they will have it sorted out, but for now, it’s just a big, and often expensive, mystery.

Update: I recently found that Google will allow you to change your email address as long as it is not an @gmail.com address. So the moral of the story is never ever sign up for AdSense with an @gmail.com address.

Quick Facts

Featured Story

after Firefox 3.6 – new font control features for designers

What

Video and article showing advanced typography features coming to CSS

Who

Jonathan Kew, John Daggett

Excerpt

Using @font-face allows web designers a wide palette of font choices and with commercial font vendors supporting the WOFF font format, the set of font choices should improve even more. So the next step is clearly to try and make better use of features available in these fonts.

For many years, “smart” font formats such as OpenType and AAT have provided font designers ways of including a rich set of variations in their fonts, from ligatures and swashes to small caps and tabular figures. The OpenType specification describes these features, identifying each with a unique feature tag. But these have typically only been available to those using professional publishing applications such as Adobe InDesign. Firefox currently renders using font defaults; it would be much more interesting to provide web authors with a way of controlling these font features via CSS.

Håkon Wium Lie of Opera, based on discussions with Tal Leming and Adam Twardoch, proposed extending the CSS ‘font-variant’ property to include values for font features. Mozilla is actively working on a new proposal along these lines. This is a fairly big addition to CSS, so it will most likely involve some complex discussions about how best to support these features.

Comment

I’m so glad someone is working on this. The type community has embraced ligatures as far back as printing began. But the Microsoft powered browser team would never be able to appreciate or in some cases, even understand these features, and they certainly wouldn’t be the first to implement them.

When I worked there, there were people who appreciated type to some extent. The people who design all those fonts Microsoft keeps unleashing onto the world. Like everyone at Microsoft, though, can they really appreciate type very much to work in such a non-creative environment?

There still isn’t broad OS level support for ligatures, and Word has some very limited Office only implementation. That’s the funny thing about Microsoft, I keep expecting someone to say, “We have the most used word processor in the world and we don’t support this obvious feature.” But they never do. I worked on WPF, and that has support for a bunch of typography features, (tested using Zapfino) but I would highly recommend against using that. Microsoft won’t even use it.

Quick Facts

Featured Story

Five questions with Russell Brown

Who

Russell Brown

Excerpt

Russell Brown (RB): There is no Adobe Photoshop CS5! Russell Brown (referring to himself in third person in jest) gave a technical presentation showing technologies that may or may not appear in the future. I do not know anything about CS5. What was shown online–Patch Matching, Wet Paint and Warp–are technologies which I had demonstrated at Las Vegas a few weeks ago.

But with regard to Patch Matching, it’s a miracle. It’s the next miracle. From what I’ve seen in Adobe’s lab, it’s going to be outstanding. We are making it possible to correct things so well, like getting rid of telephone or power lines going through photos, and this solves a common problem that photographers face. It’s a cool, easy-to-use technology which Adobe is working on.

Comment

Here’s an interesting short interview that will at least give you pretty much the most up to date current information Adobe is willing to give about Photoshop CS5.

In the interview Russel emphasizes that Photoshop is a tool for professionals, and I completely agree. I talk to so many amateur photographers, like myself, who immediately want to “learn” Photoshop. I always say, “Trust me, you don’t!” Then add, “If and when the features in iPhoto aren’t good enough for you, then maybe you should get Photoshop.” I do almost all of the retouching of all my photos in iPhoto. It does everything I want and saves me huge amounts of time over trying to make the same adjustments in Photoshop. I use Photoshop to create images, and many of its tools are completely necessary for me to do that. I always try to point out that while I have the skills to use Photoshop to retouch my photos and I still choose not to, even though I use Photoshop every day.

Yes, its a cutsie name, but The Bug Genie is a PHP bug tracking system that is the opposite of Trac in the best possible ways.

First of all you create a MySQL database and copy The Bug Genie files to your web server. Then you go to your website, where The Web Genie greets you with a short setup. At the end, it gives you an administrator login and password and you’re ready to go.

It’s that simple to set up. No command line required.

The help is built right into the system in the form of little question marks strategically placed that you can click on. They usually answer any beginner questions you’ll have.

How to create users? Well you can do it quite easily, but why bother? Just let your users sign themselves up with their email address and create an account. Just like WordPress.

It’s very powerful and has Subversion integration. Actually, for my purposes, it’s a bit too powerful for the minor issue tracking I need, but after having worked at Microsoft and using their bug tracking, The Bug Genie seems better. It’s certainly prettier, and prettier than everything else out there as well.

I predict great things for The Bug Genie for a lot of great reasons, but maybe for no other reason than nobody can set up any of the other bug trackers without dedicating their life to them.

]]> http://designoplasty.com/2009/09/08/forget-trac-the-bug-genie-is-awesome/feed/ 0 http://designoplasty.com/2009/09/07/trac-bug-tracking-software-is-junk/?utm_source=rss&utm_medium=rss&utm_campaign=trac-bug-tracking-software-is-junk http://designoplasty.com/2009/09/07/trac-bug-tracking-software-is-junk/#comments Tue, 08 Sep 2009 00:30:46 +0000 Matt http://designoplasty.com/?p=952

I have just spent many hours of my life I will never get back trying to get the Trac bug tracking software set up. This software is truly junk.

It’s not just that I can’t get it work, it’s that it is written as best it can to make users not be able to use it. Clearly some people get it to work, and I imagine this takes many man hours to accomplish.

I’m even using Trac auto installed by Dreamhost, so I don’t even have to think about unpacking things or anything like that, I should have a Trac that’s just ready to go, but no, there’s no such thing.

First thing I want to do is login with my first username and password, right. Well I don’t have a first username and password. And when I click the login link, it says something like this Trac is not set up for user authentication.

Now, isn’t that just screaming to people “don’t use me!”? I mean, isn’t that the first thing everyone would want to do? Log in? Create users? But not with Trac, you have to prove to them that you’re worthy of using Trac. You do this by following many complex methods of manually creating a password file and then using various other methods of begging Trac to use it. This means configuring Apache and Trac, and if you’re lucky, various other Trac plugins.

And if you thought getting Trac to work was hard, just wait till you try to install a plugin! That’s another thing they’ve royally messed up.

I never got it to work. I never got a single user or password and never saw a login page. I also never saw a place where I might enter a bug, and remember, that was the whole purpose of doing this.

At this point I can say without doubt that Trac is not for you. It’s not for anyone. Don’t encourage them and don’t use the software. When you can log in (through the web) immediately after installing and create users, then maybe. But right now, it’s just junk. Trac is just a joke all the Trac developers play on their victims, don’t fall for it.

Don’t do it!

Once you turn it on, there’s no getting rid of it. It’s a bug, but unselecting the check box will no do it.

At some point I also went into Preferences > Keyboard > Keyboard Shortcuts and pressed the “Restore Defaults” button because it said there were some conflicts. I don’t have any special keyboard shortcuts set up, so I just wanted the defaults. But now check out these two screenshots.

Command-Space now means “Select the previous input source.”

Control-Space now means “Show Spotlight search field.”

Huh? For my entire Spotlight history Command-Space has brought up the search field. But now it’s changed to Control-Space because of input sources? These screen shots show I’m not hallucinating, but can someone please commiserate with me or at least confirm that this is bizarre?

When I bought a MacBook a while ago, I got an HP C4580 printer/scanner for free. I didn’t really need a printer, but the idea of a wireless scanner enticed me. Well, once you get the printer out of the box and start reading HP’s instructions, you quickly just give up on ever even trying to set it up. The instructions are so vague, that I wasn’t sure that I would even be able to scan wirelessly if I got it set up correctly. So the printer has been just sitting there ever since I got my laptop.

After I read that Snow Leopard would download printer drivers automatically, I decided to plug the printer into my laptop just to see. It did recognize the printer immediately, but since I have no need for clunky wired printing I just looked around for a way to set up the wireless. There wasn’t a way that was immediately apparent.

I guess for some reason I decided to do a Google search to see if there was a way I just wasn’t seeing. It turns out there was, but you never even need to use the USB cable. The way is shown in this video.

I followed these instructions and never downloaded a driver or installed any HP software. In fact some instructions I read said “please do not try to install any printer software you may have.” Next thing you know I had the printer set up to print and scan wirelessly from two computers. No clunky HP software installed!

The scanning in Snow Leopard is truly awesome. Wireless makes it so much better. I always said I would never scan, it’s too tedious. But now, I might just scan some photos in. It’s not painful, for the first time in my entire life the process of scanning photos is not painful. Good job Snow Leopard!