Remember how, last week, we set up accepting invites sent to your non-Google address from your Google address? No? What’s that, you don’t have an Alternate email address section under Reminders and Notifications? Well, here’s the thing, you need to actually set up an alternate address for your Google account. That is easy enough, though; here’s how:
- Sign in to your Google Account
- Click your avatar, then Account
- Click the Security header
- Click the Edit-link next to your recovery email address
- At the bottom right, click the blue “Add an alternate email address“-button
- Add the email address, then click the confirmation link when it arrives in your inbox
That’s it; you’re done!
Here’s the scenario: You have set up your main email address to forward to your GMail address. Whenever you get calendar invites, you are unable to accept them, because they have been sent to your main email address, and not your GMail address. If that sounds familiar, I have good news: this can be fixed, and the fix is fairly straightforward. Here’s what you do:
- Log on to your Google Account
- Go to Google Calendar
- Click the Gear icon and choose Settings
- Select the Calendars tab
- Click Reminders and Notifications next to your primary calendar
- Under Alternate email address, check the box next to “Allow me to respond to event invitations forwarded from these addresses. My attendance response will come from firstname.lastname@example.org“
That should do it.
A while back, I had a user call in and tell me that he was editing a document in SPSS, and went to save his progress to the original file, when he got the following error:
The SAVE command has succeeded. However, due to contention for the specified file, the data have been saved to a file with a different name. Saved to: location\filename.sav
He further reported that the file specified was the one he was working on, and the file size had been reduced to 0kB. Looking into it, I found that this is a challenge for SPSS users for years, and is most likely caused by SPSS locking the original file, and in turn being unable to save to the same file. Knowing this, a workaround presents itself: save your work as a new file, then close and rename. Cumbersome? Sure. It will, however, prevent you from losing data.
The single most handy tool for any Linux administrator is SSH. Short for Secure SHell, it is a way to connect to another computer, using the terminal. It is immensely powerful, and gaining confidence with it can be a bit of a challenge. Before you can do ANYTHING else, you need to know a few basic commands:
- ssh user@server
- ssh -p #### user@server
Now, that wasn’t so hard, was it? But, wait a minute; what are we actually doing? The first command tells the terminal to connect, using SSH, to a server, identifying as a user (obviously, you would need to replace user with your username and server with the address of the server). The second command does the same thing, but specifies which port to use (and yes, you need to make the same replacements as before, as well as replacing #### with the port number you are trying to access).
Having logged in, we can perform a number of tasks, such as setting folder permissions, creating users, running tests on the network, and much, much, much more. This, though, connecting and logging in, is where it all starts. Without that, nothing that you know how to do makes any difference.
The first post to this blog was made on September 14th, 2007. Since then, seven years have gone by, I have moved twice, changed jobs as many times, gotten married and become a father, and embarked on a bachelor’s degree. The blog has gone from being hosted on Blogger to being hosted on WordPress, and I have moved it from a subdomain to the primary domain where you find it now.
By no means the first online presence I have had (the first was hosted by IBM back in 1996, the second was on now defunct GeoCities), this blog is the one over which I have had the most control, from which I have learned the most, and which has offered the most permanence.
This is the four hundredth entry I have made. They range from my forays into Linux to the sunk cost fallacy, and everything in between (and a lot of things not connected at all. They all belong here due to the simple fact that I say they do.
I have no plans to stop writing now. I have far too much still to cover, such as my forays into Linux and OBIEE. The topics will continue to be varied, and other topics to come are book reviews, change management and really, whatever else I find the time and inspiration to write about.
I like having my Mac start a set of programs each time I boot it, yet there are others which, under normal circumstances, would be set to load on boot that I only want to be running when I want them to be. Luckily, deciding what programs to start on boot – or not – is a simple enough proposition. Here’s how:
- Open System Preferences, then Users & Groups
- Go to the Login Items tab
- Click the lock to make changes, and input your admin password when prompted
- To remove programs: Highlight the programs you don’t want to load, and click the minus sign to remove them
- To add programs: Click the plus sign, and navigate to the program in question
That’s all there’s to it.
Business Intelligence, at its core, is about taking data from different sources, compiling the data into a format that can be used, and running reports on that data. The process of compiling the data is called extract, transform load; ETL for short. If you are going to work with Business Intelligence, you need to have, at the very least, a rudimentary understanding of ETL. Here’s a quick overview:
Extract means connecting to all the data sources you have, such as payroll, accounting, inventory, sales etc. (basically any system or data source that can provide you with base data for statistic analysis), and extracting the raw data.
Transform means adapting the data from their disparate formats to one where the formats are the same between datasets, to fit the needs of the business.
Load means loading the transformed data into a system designed to run reports on the data, such as a data mart or a data warehouse.
Once the data has gone through ETL, you can either run analysis to get reports from the data, or run it through ETL again to load it into a different data store.
Here it is. The reason I chose Oracle Linux over all the other distros out there. Over Debian-based Ubuntu, with which I am somewhat familiar, and over Fedora-based Red Hat, which is commonly used in enterprise environments. The reason I chose Oracle Linux, is that I want to learn more about Oracle Business Intelligence Enterprise Edition, OBIEE.
The learning curve with configuring OBIEE is significant, to say the least, which is why I recommend running it under Oracle Linux on a Virtual Machine, snapshotted at regular intervals. That way, when you run into a roadblock, you can make a note of what went wrong, revert to the last snapshot and avoid that particular roadblock the next time around.
Over the coming weeks and months, I will be chronicling my journey through getting OBIEE set up. Don’t worry; there’ll be other stuff in there as well. For now, here are a few links that I have found useful:
Before starting out with Oracle Linux this summer, my experience with Linux had been more or less confined to Ubuntu, which stems from Debian. Oracle Linux, on the other hand, stems from the Fedora project and Red Hat. Though they both build on the same kernel, they diverge from each other in a few important aspects. Software available in the repositories for Debian are divided into free, non-free and contrib. All software available in Fedora’s repositories are free.
To me, the single most significant differences between Debian and Fedora (and by extension between Ubuntu and Oracle Linux) is how software is distributed. Anyone who has used Ubuntu knows the apt-get command. Try that in Oracle Linux, and you get an error. Apt-get simply doesn’t exist. The reason is that Oracle Linux doesn’t use the apt-get dependency resolver, but rather one called yum. Add to that different formats (deb and rpm), and package managers (dpkg and RPM), and you start to see the level of difference. For a list of yum commands, have a look here.
There are other differences, too, I’m sure, but none are as important for basic use and understanding of operation of the OS as the ability to install software.
In order to keep the virtual harddrive on my VM as small as possible, I prefer keeping installers on a network share that I can connect to through FTP. The terminal command for connecting to FTP servers handily enough is ftp. When running that command, terminal returned “Command not found”. It turns out that FTP is not installed on Oracle Linux by default.
Installing it is easy enough, though. In terminal, as root, run the command
yum install ftp. Answer yes to all questions and hey presto; FTP is installed. The lesson here: Never assume a resource is installed. I’m sure I’ll run into more of these down the road, and it’s a good lesson to have learned, thought I suspect it is one that I will be reminded of time and again.