As part of the DST changes, this weekend I had to upgrade our phone switch to the latest version. This went fairly smoothly, with the exception of them changing the default passwords, and me not catching it at first. When I got to updating the CCC, things didn’t go too smoothly. At first, I’d removed the old IP Office Manager software from the server. This was because it hadn’t been updated in a while, and was out of sync with the main server now, and I didn’t believe it was being used. Nothing in the upgrade/install docs said it’d be needed either. Then I uninstalled the CCC software. This is where things didn’t start quite as they should have.

Installing the new version was relatively easy. Just follow the install steps, or the upgrade docs. Then it came to testing it. I first tried to get access to the “Delta Server”. This worked, however it said it wasn’t connected, and couldn’t get licenses. This reminded me of a post I’d made recently, and thought they might be tied together. So I decided to ensure the “Key Manager” service was started. To my horror, there was no key manager service. So something must have gone missing in the install. I reran it twice, and nothing appeared. I did some googling, and nothing really popped out as the solution. So I decided to take a look at the other server, and sure enough, the key manager service was on there. So I grabbed the admin utilities, and started up the install on the CCC server, and there it was as part of the install. So I reinstalled the admin utilities on the CCC server, and then restarted the delta service. I then forced it to reconnect to the phone switch, and the Archiver database. All seemed to be well.

This is where more stuff appeared to be broken. I had recently made modifications to our call menus and wanted to increase the reporting a little to allow us to get a better insite into what is going on, and where the calls are going. So I enabled some reporting options, and gave it about 20 minutes before trying to load a report. When I attempted to open the report, I was promptly thrown an error:

Load Report Failed

This wasn’t very descriptive, nor were the logs, so I started to take a look around. This is where I stumbled across this item from the creators of Crystal Reports (this is what the CCC reports are based on). While the article was specifically about Windows 98/ME, it did have the same error, and could be the possible culprit. So I copied the line:

> cd "Program Files\Common Files\"
> cd "Crystal Decisions"
> regsvr32 "1.0\bin\CRQE.dll"

This generated an error that the file didn’t exist. This turns out to be caused by the fact that we have 2.0 installed, and not 1.0. So I adjusted the directory structure, and executed. I proceeded to test a report… This time with success. Apparently during my install/reinstall/uninstall fiascos, somehow the crqe.dll file had not been correctly registered.

I’ve been pretty good at keeping my server up to date. I’ve taken a general policy of “if it isn’t broke, don’t touch it”. It works well, for the most part. When it comes to servers though, security is a good idea, and that plays havoc with the previous mantra.

Warning: This post is long… not only is it long, but I’m pretty sure it’ll be filled with profanities, and general obscene guestures, and words… Read on if you dare…

We have a weird issue at work where an import task will occasionally hang waiting for nothing. This can sometimes be harmless, and othertimes disastrous. When it’s disastrous, it causes the main table in our database to be locked, denying other services/users from performing any inserts, or updates, or even selects. This is a big problem… obviously.

Recently I’ve felt this urge not to work at all. I’m not sure why. I think it’s just a case of procrastination. I have plenty of things to do, though some require after hours work (switch upgrades, new rack PDUs, etc) which I normally enjoy doing. However, I’ve not been feeling too appreciated at work recently, and its partially coming from the fact that our dev/qa environments were moved to the corporate environment. While this should make me happy as it’s one less thing to worry about, in the last two weeks, we’ve had about 6 or 7 outages, and a constant slowness. I’ve practically had my hands tied on resolving the issue as I have no access to any of the corporate networking equipment.

I’ve had several people comment on the stability of the two environments since they went to corporate compared to when I was managing them. I think I had maybe 2 or 3 outages in 2 years (both related to RAID controller failure). That makes me feel good in the fact it shows people do notice the work I do, even if they don’t initially notice it. It also reminds me that I am now unable to diagnose any issues, let alone attempt to fix them.

I guess it’s less procrastination, and more rebellion to the fact that I have the ability to help, I’m just entirely cut off from being able to help.

Being pretty much stuck on Windows due to the tools I have to use (Outlook, Visual Studio, SQL Enterprise Manager), I find that developing home stuff a little bit of a challenge. Sometimes I’ll SSH into one of my development boxes, and work from there, but it becomes a bit of a strain jumping around code. So I turned to Eclipse. It’s a massively extendable editor.

The problem comes when trying to commit code changes for some of the projects I’m on to SourceForge. SourceForge use CVS over SSH. This requires a secure connection to the server before the CVS connection can be established. This is a relatively easy task with the help of putty, and these instructions by Stefan Wehr. The instructions are a little out of date, but easy enough to adapt to environmental changes. For example, the SourceForge CVS names are now project.cvs.sourceforge.net.

I can now manage all of my projects on Windows, and if I finally get the chance to get off of it, I can simply move my repositories.

Enjoy…

This year, all chaos and mayhem is going to break lose. Well, at least that’s the theory. This year the Government decided to bump the DST from April to March. This isn’t too bad, until you start to think of the impact of all the computers having the wrong time. Fortunately, our fearless developers have produced updates. Microsoft have released a number of updates (details here), though a full patch hasn’t been released for Windows 2000, there is a reg hack.

Then there is linux. Fortunately enough, two of my boxes have already been updated. Those are driven by Debian, and probably got updated as part of a security update that was installed. The unfortunate thing is that two of my boxes are not updated. This is because they are Gentoo, and I have only been deploying security software updates. This didn’t include updates to things like GLIBC which is where the time libraries are deployed from. However, some time during the last 6 months or so, the Gentoo team decided to split time management out of the glibc libraries into their own ebuild called sys-libs/timezone-data. This is good for issues like this, however from a mention on the Gentoo Forums, it may require a certain glibc version that I probably don’t have. So looking around, I found that you can actually do the updates manually. I stumbled across this little gem. The unfortunately issue with this is that the zdump, and zic applications are bundled as part of the sys-libs/timezone-data package. You have to love circular dependencies. Off to find a solution…

I hate it when people don’t like what I tell them.

We are currently having an issue with some of our servers caused by a core switch. As they are not entirely sure what is wrong with it, they have not set an ETA, which I passed onto the parties affected by this outage. The problem is, they evidently didn’t believe me, and decided to ask other people. Why do people insist on doing that? It’s incredibly annoying, and makes it look bad on me.

This isn’t the only thing either. I’ve told people in the past that something wasn’t possible for a number of reasons, only to be CC’d in an email at a later point in a conversation with somebody else, as they decided to ask them instead.

Well, I took a leap, and upgraded to WP2.1. Quite a smooth upgrade. Of course, I demo’d the upgrade on my dev box at home, and found the only plugin I had to upgrade was the markdown plugin as it generated preg_replace errors. Once upgraded, all was smooth sailing again.

Some people cannot drive. I had somebody reverse into my car whilst driving out of the carpark yesterday. Fortunately no bodywork was damaged. Just got to get my alignment fixed… again as she hit my wheel as I was turning to avoid her.

On a lighter note compared to my last two…

Some people are like a Slinky. They really have no purpose, but you can’t but help smile when you push them down the stairs

Ahhh, now I feel better.