About Coding, Music, Family and anything else I want
I wrote a quick utility to help me figure out the available IP Address range when given a specific IPv4 address and slash number. You know something that looks like this: 192.168.0.0/16. The utility should tell you the range of IP addresses that covers.
I’m not saying I’m a master of CIDR Slash Notation or Subnet masks… but this seems to be providing good results 🙂
Over the weekend I spent some time setting up GitLab on Ubuntu Server using a WiFi card. For those of you who do not know what GitLab is, check it out. I stumbled upon GitLab several years ago when I was looking for a self-hosted GitHub alternative. Since then, GitLab has greatly improved, and setting it up is fairly easy.
First, you are going to want to obtain the Ubuntu Server install. You can download this from the Ubuntu Website.
The second step I took was to find an old desktop I wasn’t using anymore. This is going to be my server. I installed a PCI-E WiFi card in the sucker, because, honestly I’m too lazy to run the network cable.
Note: I tried to setup the server multiple times using just the WiFi card. I wouldn’t recommend it as it was a very frustrating process. I’d highly recommend hooking your new server up via an Ethernet cable at least until you setup the WiFi. It’s far easier and saves a ton of time.
After I hooked up my server with the Ethernet cable I booted to the Ubuntu Install disc and began the installation process. The process itself is really quite simple. There are a few questions you have to answer but the whole thing should be over in less than 30 minutes. I just overwrote everything on the hard disk. After it’s done installing it’s going to ask you to remove the installation media. At that point it should reboot, load up, and show the login screen.
Note: Ubuntu Server does not come with a GUI. Everything is done via the command line. You can install a GUI if you want, but there isn’t a GUIÂ packaged in.
Now that I had Ubuntu Server installed I went ahead and logged in. The first thing it showed me was that there were some updates to be installed. So I ran the following commands to update the system:
sudo apt-get update && sudo apt-get upgrade
Now that the system was updated I wanted to get the WiFi working. In order to get the WiFi working I used nmcli. nmcli is a command line tool that comes with the Network Manager package. Some people might not like using this tool because I believe it installs some GUI dependencies. Honestly, nmcli was the easiest method I found to get the WiFi working, so I don’t really care about the small amount of dependencies that the Network Manager package comes with.
sudo apt install network-manager
Alright. I had the network-manager package installed. Now to connect to my WiFi network.
I read through the “man” page for nmcli. It looks like I can get a list of wifi access points in the area by running the following command.
nmcli device wifi list
Yes! That actually gave me a list of WiFi access points in my area. I saw my home network listed. I was so happy to see this because it meant I didn’t need to configure anything else. The Ubuntu server had recognized my wireless card and the card was working. That made me so happy… 🙂
Next step would be to actually connect to my WiFi access point. According to the nmcli man page I can connect using nmcli device wifi connect. My access point requires a key, and it looks like nmcli supports connecting to an access point with a key… so this is a good thing.
nmcli device wifi connect MyAccessPoint password 123456789ACB
Boom! I ran that sucker and it actually worked! I had been struggling and struggling with this before – nmcli is like my new favorite thing ever. EVER.
At this point I rebooted the server and disconnected the Ethernet cable. I wanted to see if the server would automatically connect to the WiFi access point on boot. It seemed to take a long time to start. After it started up I logged in. I tried to ping google.com. No dice. I waited a few moments and tried again… it worked!
I made sure OpenSSH was installed on the server so that I could manage it from another computer.
My WiFi was now working on the Ubuntu Server. It was connected to my home network, and it automatically connected after the server was turned on.
Now that I had the WiFi connected I wanted to get GitLab all setup. Luckily, the folks at GitLab have made this incredibly easy. They have a great guide setup here. There are really only a few commands you need to run and then you are good to go. Let’s go ahead and list those commands really quick.
sudo apt-get install curl openssh-server ca-certificates postfix
These are things that GitLab needs in order to run successfully.
curl -sS https://packages.gitlab.com/install/repositories/gitlab/gitlab-ce/script.deb.sh | sudo bash sudo apt-get install gitlab-ce
I used the above command. However, GitLab mentions an alternative if you aren’t comfortable with a piped script. You can find the alternative on their guide page.
sudo gitlab-ctl reconfigure
The people behind GitLab have really made it incredibly easy to get this up and running. Now you just need to login to your new GitLab server. When you first visit the page you will be asked to create a new password. This password can be used in conjunction with the “root” username to login to the system.
That should be it. Your GitLab server is setup and working. You’ve setup the WiFi card so the server is connected to your network. You’ve got OpenSSH installed so you can manage the server from another machine. You’ve installed GitLab so you can host your own internal Git repositories (as well as collaborate with others on your team etc…).
The last things I would do:
Alright – that’s it. This post ended up being a bit longer than I thought, however I’m glad I’ve got it documented it. (Even if there wasn’t much to document).
I use the GIT command line interface a lot. It helps me keep my Git repositories looking sharp and clean. Interactive rebase auto-squashing with posh-git+ConEmu ftw!
However, from time to time I will notice that the GIT cli is asking me for my SSH RSA passphrase more often than I’d like. Sometimes it even asks on every pull. That’s annoying.
It is possible, however, to only enter your passphrase once per session. Setting this up should be as simple as doing the following:
ssh-agent in your powershell environment.ssh-agent
ssh-add
Excellent! That should be it. Now you should be able to push, pull all you want without having to insert your passphrase more than once per session.
So I found this amazing post that has helped me a ton:
I am constantly referring to that post in order to get the correct commands for launching something (like ManStu) as a domain user.
So, I figured I would post the command here just in case that post goes away (as posts on the internet tend to do).
The command is:
runas /netonly /user:domain\username <program you want to run>
Example:
runas /netonly /user:domain\username "C:\Program Files (x86)\Microsoft SQL Server\120\Tools\Binn\ManagementStudio\Ssms.exe"
Slack is awesome. I love it. You love it. Everyone loves it. If you don’t love it than that is probably because you haven’t heard about it. So.. go hear about it. Yep – now you love it too. You’re welcome.
One of the cool things about Slack is that it archives all of your conversations. So you can refer to (and search through) previous conversations you’ve had. This morning I was going to do just that, look up an old conversation I had with a previous team member. However, I couldn’t find his name anywhere via the normal slack interface. Because I couldn’t find his name I figured that he must’ve been deleted and my conversations were lost forever… :sad_panda:
But – don’t despair! As it turns out, you can’t actually delete a user in Slack. You can, however, disable a user. And the cool thing about disabled users is that you can still access the messages you’ve sent to them.
Accessing the messages is simple. Find the People & user groups link in the Slack interface. If you cannot see it then you might have to click the â‹® More link. On the People & user groups page find and open the Filter menu. Click on the checkbox to Show deactivated accounts. Find the user you want, click on their portrait, and finally, click the Message icon. This will bring up the Message Archives for that user.
Boom sauce. You can now view your archived direct messages from previous members of your team.
It is also possible to access messages from a disabled user using the Slack search functionality. In order to do this you can simply enter from:@userName (where username is the name of the user you are looking for) in the Slack search bar. The search results should actually show you messages from the disabled user.
Note: Keep in mind. If you are using Free Slack then you are limited to 10k messages. It’s possible that the messages you wish to find are beyond the 10k limit.
When using IISExpress to develop a Website project it is nice to have a specific machine name and port to refer to.
You can specify the machine name and port by editing the binding in the IISExpress applicationhost.config file. The following information was gleaned from this answer on StackOverflow.
- Open your applicationhost.config file. It most probably will be
%userprofile%\Documents\IISExpress\config\applicationhost.config, but inspect the output fromiisexpress.exeto be sure.- Locate your WebSite entry and add following binding with your machine name.
<binding protocol="http" bindingInformation=":50333:your-machine-name" />- Restart IIS Express
I actually specified my machine name as well as the local domain.
bindingInformation=":8013:my-machine-name.my-domain.local"
This seems to be working just fine for me. If you are wanting to actually share that url for others to access (maybe within your own work network) you might have to run this in a administrator command prompt as specified in this StackOverflow answer:
netsh http add urlacl url=http://vaidesg:8080/ user=everyone
I’ve recently noticed a tiny little button show up in the top right hand corner of Chrome next to the “Close”, “Maximize” and “Minimize” buttons. This button is for Google Chromes newish “Avatar Menu”.
It looks like this:
Since I don’t typically like new things.. I immediately went looking for a way to make it go away.
That said, you can use Chrome flags to turn off the avatar menu in Google Chrome.
Kaboosh. The new Avatar menu is gone (for now).
The Windows 8.1 experience that I’ve shared below is just that, an experience. Windows 8.1 itself is fine, I’d prefer if it gave me more customization options. Personally, I’d prefer not having a Windows button (I’ve got a windows key on my keyboard). Personally I’d prefer full screen search over the tiny search bar in 8.1. So Microsoft would’ve done better to provide personalization options, not to choose for you.
And I’m a little ticked off.
First off, my computer updated itself without my knowledge nor my permission. It had asked me a few times, to which my answer was always “Not Right Now”. However, it decided, of it’s own accord, that it would update itself. That kind of behavior is not acceptable.
In addition to updating itself to Windows 8.1 it also decided that it would be helpful and download/setup/install a bunch of apps for me. Once again it did so without asking for my permission. It then decided that it wanted me to create a special Microsoft Account to use my personal computer, Luckily, I found a way around that (using my smartphone, because my computer wouldn’t let me use it).
When I finally got into the computer I discovered that it had changed a number of things that I did not tell it to.
All in all, I wasn’t that happy with the way Windows 8.1 decided to force itself upon my machine.
This whole process would’ve been a whole lot less frustrating if it allowed for more input from the user. It should’ve told me about all the changes it wanted to do (add IE, add the Windows button, modify the Startmenu), and more importantly, it should’ve given me the option to keep things the way they were.
So… now I proceed to search the internet for ways to make my computer the way I like it again. Which makes me even more frustrated.
**UPDATE**
It appears there is no built in way to turn off the Windows button in the taskbar. There is also no way to make my Start Screen searches full screen by default anymore. (Ok, so if I actually click the apps view, and then search, the search is the way I want it to work). The only way to get these things back to the way I like them is by re-installing windows 8, which is pretty annoying. All in all I’m really sad with the way this whole update happened.
Kaira mentioned that she’s been postponing the update by doing the following. Maybe that will help some of you.
What I do is, when the prompt asking to restart at a time of my choosing sets, set it to as late as it goes, then rush over to the Windows Store. Click Updates (top right) and you’ll see it downloading 8.1. Click that and something at the bottom comes up, and you can select Cancel.
We’ve been using SCRUM at work. It’s certainly been a learning process. During that process I’ve learned that I’m not a very big fan of SCRUM. I’ll write more on that later. That said, I’ve come up with a few ideas which I hope will make my time, and my teams time, with SCRUM a bit better.
0, 0.5, 1, 2, 3, 5, 8, 13, 20, 40, 100, INFINITY
When we see numbers like 13 we immediately think that they are “big” numbers. So when we estimate we tend to pick numbers under 13 even though the story being estimated might require a 13. So, I say pick a number between 1 and 10 in your head and then translate that to the SCRUM numbers.
| 0.5 | 1 |
| 1 | 2 |
| 2 | 3 |
| 3 | 4 |
| 5 | 5 |
| 8 | 6 |
| 13 | 7 |
| 20 | 8 |
| 40 | 9 |
| 100 | 10 |
I really hope these things help. Ideally I’d like to move away from SCRUM. However, if I’m going to have to use SCRUM then I might as well figure out how to make it better.
Do you have any ideas on how to improve the SCRUM process? What do you do? How did your teams make things better?
Generally in .NET MVC you would use the default model binders to deserialize JSON to a .NET object. However, there are some cases, involving “complex” collections, where this becomes a bit tedious to do. Microsoft provides a few ways around this, but none are satisfactory. Unlike the default model binders, the JavaScriptSerializer provides support for deserializing JSON to a .NET object with complex collections.
Take the following simple class and method signature for example:
// simple class
public class SaveInformation
{
public string Name { get; set; }
public Dictionary<string, List<SaveItem>> Components { get; set; }
}
// method signature
public ActionResult Save(SaveInformation saveInformation)
Below I’ve included some simple JSON that we will pass to .NET via AJAX:
{"SaveToSession":false,"Name":"My Stupendous Thing","Components":{"Component1":[{"ProductId":"1234"}]}}
We are passing the JSON to .NET using AJAX with the “application/json” contentType. Because we are using the “application/json” contentType, .NET will automatically call the JsonValueProviderFactory and map the information over to our SaveInformation model. It seems pretty straight forward, the Save method will receive a SaveInformation object when it is called.
Not so straightforward unfortunately, the Save method does receive a SaveInformation object. If we inspect our SaveInformation object we see the Name is populated perfectly fine, but the Components dictionary ends up being null.
The reason for this? It seems that the JsonValueProviderFactory doesn’t fully support pure JSON syntax. In order for .NET to properly parse a complex collection, you actually have to give it a numerical index.
From http://msdn.microsoft.com/en-us/magazine/hh781022.aspx:
Though it’s somewhat counterintuitive, JSON requests have the same requirements—they, too, must adhere to the form post naming syntax. Take, for example, the JSON payload for the previous UnitPrice collection. The pure JSON array syntax for this data would be represented as:
[ { "Code": "USD", "Amount": 100.00 }, { "Code": "EUR", "Amount": 73.64 } ]However, the default value providers and model binders require the data to be represented as a JSON form post:
{ "UnitPrice[0].Code": "USD", "UnitPrice[0].Amount": 100.00, "UnitPrice[1].Code": "EUR", "UnitPrice[1].Amount": 73.64 }
I know right? It was hard for me to read this as well due to the anime-sized tears in my normal-sized eyes. The problem is, no browser I know of has a JSON.serializeForDotNet function. So this leaves us with two options, make our own JSON serializer, or make our own ValueProvider/ModelBinder. Neither of these options sounded very appealing to me, so I went with the hidden third option. Pass in a string 🙂
You can use the JavaScriptSerializer class, located in System.Web.Script.Serialization, to deserialize JSON information and, as an added bonus, it actually handles properly formatted pure JSON!
So, the new method looks like this. We pass in a string value and deserialize it using the JavaScriptSerializer. Now we receive a SaveInformation object with a fully populated Components dictionary.
public ActionResult Save(string saveInformation)
{
// instantiate a JavaScriptSerializer
JavaScriptSerializer serializer = new JavaScriptSerializer();
// use the JavaScriptSerializer to deserialize our json into the expected object
SaveInformation saveInformationObject = serializer.Deserialize<SaveInformation>(saveInformation);