A Git Primer for Sysadmins

August 30, 2013August 31, 2015 Josh Reichardt 8 Comments

I am writing this post with the intention of giving readers (and other sysadmins) that are unfamiliar with Git a brief introduction to get up and running as quickly as possible. There are numerous guides out there but none I have found so far with the spin of getting things working quickly for system administrators. I recently went through the process of submitting some work of mine to an Open Source project and went through all of the work myself and thought I would write up a little guide to getting started, just in case anyone is interested. Because of this experience, I thought the process could be streamlined if I presented a simple set of steps and examples for other sysadmins so that they can get started and get comfortable using Git as quickly as possible. By no means am I advocating that I am a Git expert, I simply know enough to “do stuff” so take this guide for what it is, a simple introduction. Hopefully it is enough to get the ball rolling and get you hooked on revision control!

Step 1: If you haven’t already, head over to github and set yourself up with a free account. The steps are trivial, just like setting up any other account. You will need to install Git on you machine locally so that you can do everything, sudo apt-get install git in Ubuntu, or, with a little more legwork if you are using Windows.

Step 2: Once you are all set up and running you can either create a new repository or fork off of somebody else’s repo, which will essentially make a copy of their work to your user account in github. Since I was contributing to a public project I forked their repo. This is typically (for beginners at least) done through the website by browsing to the project you are interested in and then clicking the fork button at the top right.

Once you have copied or forked the project you want, you can begin working on making your own changes and updates to the project. This is where much of the initial confusion came in for me. I should make a note here; there are a few things that must be done through the website and others that should be done with the git tool you downloaded and installed earlier.

Step 3: Clone the repo down to your machine locally to begin making changes. This is done via command line with the following:

git clone https://github.com/$username/repository.git
git clone https://github.com/jmreicha/curriculum.git

Once you have the repo cloned down to your machine, let’s create a branch which we will use to work on the specific piece of the project we are working on. There are a few more common commands that are used for working with repos. In some scenarios the public project may change or get updated while you are working on it so you might want to pull in the most recent changes to your own fork. The following commands are used to work with public repos and forks.

Fetch new changes from original repo:

git fetch upstream

Merge the fetched changes into your working files:

git merge upstream/master

Pull newest updates and changes from a for or repo:

git pull

Okay, let’s assume that our repository is all up to date and we are ready to start working on the new changes that we want to make. First, we need to create a local branch to make the changes on. The command for creating a branch is:

git checkout -b $topic
git checkout -b my_topic

Where $topic is the specific change you will be working on. Then go ahead and make your changes and/or updates with your favorite text editor. If you have multiple changes to make it may be a good idea to create a few different branches using the command above. To switch between branches use the following:

git checkout $branchname
git checkout my_topic

If you are working on a branch and either don’t want to commit the changes or the branch becomes obsolete, you can delete the branch with either of the following commands:

git reset --hard
git branch -d $branchname
git branch -d my_topic

Step 4: Assuming you are all done with your branches and are finished making your changes the next step is to commit the changes. This will basically record a snapshot of (the state) the files. Typically for a commit you will want to do one thing at a time to keep track of things more easily. So you will make one change and then commit and then repeat if there are numerous changes that need to be made. The command is:

git commit -am "update goes in here"

Now that we have created a branch, made our changes, written our changes so that git knows about them and can keep track of them, we are ready to go ahead and merge the code that we have changed locally back up to github. Start by pushing the branch up to the github repo.

git push $remote $branchname
git push origin my_topic

Next, merge your changes. This is done by changing to the “master” branch and then issuing a merge command on the branch that was created to make changes into. So here is what this process would look like:

git checkout master
git merge $branchname
git merge my_topic

alternatively if you just want to quickly push out a change to the master branch you can issue a git commit locally and then a git push to update the master branch without merging the items first.

git commit -am "updated message"
git push

I like to use this for local repositories where I just need to get things up quickly.

Step 5: At this point everything should be accurate on your own personal github page. You can double check your commit history or look at the last update time to make sure your changes worked their way into your public fork or repo. If you are contributing to a public project the final step is to issue a pull request on the github site. Once this is done, wait to hear back from a moderator of the project if your changes need to be fixed before they accept them or wait for a response saying that your changes were good and get merged to the project.

Git, to me, was a little bit confusing at first. Like I said earlier, I still only have a generic and basic understanding but I think it is important for system administrators to know how to commit code to projects and how to keep track of their own code and projects as well. Even if the code is only small scripting projects it is still beneficial to publish since github is a public site. Potential colleagues and employers can browse through your work, which increases your visibility and can even lead to job offers.

If you’re interested in learning more, I found this site to be a great resource.

Enable telnet with PowerShell

August 18, 2013May 1, 2017 Josh Reichardt 1 Comment

Every once and awhile you will probably encounter a situation where you need to enable and then use telnet in a security focused environment. In certain situations telnet can be a great tool to test the functionality of firewall rule. Iif you aren’t certain whether or not a rule is working telnet can be a great way to help debug. The problem in Server 2008 and above is that telnet isn’t enabled by default. Luckily with PowerShell it is easy to enable the telnet functionality.

The following set of commands is a quick depiction of how you can enable telnet from a PowerShell prompt to ensure the ability of testing certain ports. Try it out.

Import-Module servermanager
Add-WindowsFeature telnet-client

Bam! As always, it is always easier to stay in command prompt and this is a great way to test port connectivity. I can understand why telnet is disabled by default on fresh server builds but sometimes it can become useful to have telnet as a tool to test connectivity. If you would like to debate the merits of disabling/enabling telnet on a server just drop me a line, I obviously will not be focusing on this aspect here. Anyway, just as easily as it is to enable telnet through PowerShell it can be disabled with the following command. If you already have the server manager module imported, skip to the second command.

Import-Module servermanager
Remove-WindowsFeature telnet-client

That’s all it takes. Very simple and very straightforward.

For more tips and tricks as well as general information about how PowerShell works, check out the venerable Learn Windows PowerShell in a Month of Lunches.

This book is one of my top recommendations on the book recommendations page, especially for learning Powershell and Windows administration.

Setting up Git in PowerShell

August 6, 2013August 31, 2015 Josh Reichardt 4 Comments

It seems like everybody is using git these days. And for most, not everybody is stuck using Windows in their day to day workflow. Unfortunately, I am. So that means it is much more painful to get up and running with a lot of the coolest and best open source projects that are offered by members of github and other online code repositories being shared via git. However, there is hope and it is possible for Windows users to join the git party. So in this post, I would like to describe just how to do that. And it should only take a few minutes if done correctly. I will mention beforehand that there are a few steps that need to be completed in order for this technique to work successfully that typically are taken care of in a Linux or OSX environment.

The goal of this post is to work through these steps as best I can to get users up and running as quickly as possible and as easily as possible, reducing the amount of confusion and fumbling around with settings. This post is designed for beginners that are just getting their feet wet with git but hopefully others can use it as a resource if they are coming from a different environment and are confused by the Windows way of doing things.

First step – Download and install the git port for Windows.

This is pretty straight forward. Download and run the executable to install git for Windows. If you just want to get up and running or are lazy, you can leave all of the defaults when you run through the installation wizard.

Second step – Add the git binaries to your system path variable.

This is the most important step, because out of the box git won’t work in your ordinary PowerShell command prompt, it needs to be opened separately. So to fix this and add all the necessary binaries open up your environmental variables (in Windows 8).

Computer -> Properties -> Advanced -> Environmental Variables

and add the following value to the PATH variable.

C:\Program Files\Git\bin

Here is what this should look like in Windows.

Third step (optional) – Download and install posh-git for better PowerShell and git integration.

I have highlighted part of this process before in an older post but will go through the steps again because it is pretty straight forward. To be able to get posh-git you need to have a sort of PowerShell package management tool called PsGet (instructions here). To get this tool run the following command from your PowerShell command prompt.

(new-object Net.WebClient).DownloadString("http://psget.net/GetPsGet.ps1") | iex

Once the command has completed you should be able to simply run this install command and be finished.

install-module posh-git

That should be it. With these simple steps you should be able to utilize git from the command line like you are accustomed to on other operating systems. As I said, there is a tad more leg work but you can really utilize the flexibility of PowerShell to get things working. I hope it helps, and as always let me know if you have any tips or questions.

Quickly Find Exchange Database Usage

July 19, 2013August 31, 2015 Josh Reichardt 6 Comments

Here is a Powershell script you can use to quickly determine the total amount of space taken up by all of your Exchange database files (edb files) on an Exchange server. I’d like to note that this may not necessarily be a 100% accurate representation but is a great way to get a ballpark number without having to add the numbers up yourself, manually.

$dbs = Get-MailboxDatabase -Status

foreach($db in $dbs) {

$edbsize = $db.DatabaseSize.Tobytes()
$totalsize += $edbsize

}

Write-Host $totalsize

I noticed that I had no way to calculate the total amount of space being used by my Exchange databases the other day. And even after scouring through teh Googles I was unable to find what I was looking for quickly so I wrote this script up quick to fix that problem. Just copy the previous bit of code into a ps1 file with notepad and execute the script from your EMS. It is a super simple way to iterate through all the databases, save their sizes to a variable and then spit that variable out when it is complete.

Disable Offline Files in Windows 7

July 17, 2013August 31, 2015 Josh Reichardt 1 Comment

Offline files in Windows are a set of features that essentially give users the ability to work with files off of or outside of the network. So for example if a user had a laptop that had a mapped drive or network share and were to take their computer outside of the network, the features offered by offline files would allow this user to continue working with these files. I will not cover the details of how all of this magic works in this post, I just want to show people the best way I found to disable this feature with the least amount of problems. If you want to go straight from the source, here is the original article the gave me about 95% of the information necessary for accomplishing this task.

The remainder of this post will detail my findings and experience from the link above. This feature (offline files) is enabled by default in Windows 7. Here is a good overview of the benefits of offline files. However, for me personally as an admin, this feature so far has caused much confusion in the work environment for users that are not accustomed to having such a feature in our move towards Windows 7.

These settings can of course be controlled on a per user basis by changing the settings and configuration of the “Sync Center” tool in Windows. But when you are involved in a larger environment and need this sort of process automated for many users, Group Policy becomes the most effective way to handle this problem. There are a few steps to get offline folders disabled correctly so I thought I would share all the pieces in case somebody runs across a similar need as I did. The first step to disable the offline file features is to adjust the following settings in Group Policy:

Computer -> Policies -> Admin Templates -> Network -> Offline files

Allow or Disallow use of the Offline Files feature: Disabled
Prohibit user configuration of Offline Files: Enabled
Sync all offline files when logging on: Disabled
Sync all offline files before logging off: Disabled
Sync offline files before suspend: Disabled
Remove “Make available offline” command: Enabled
Prevent use of Offline Files folder: Enabled

Next, we need to tell Group Policy to shut off the offline file service and disable it on all Windows machines that have the service installed (Windows XP, 7, 8 machines). To do this you will need to modify your Group Policy settings on a machine that has the service installed it already, through RSAT. This is an important step, you will not be able to find this service if you are adjusting the GP settings from a server. This service is located in the following location:

Computer Configuration -> Windows Settings -> Security Settings -> System Services

The specific service we are looking for is the “cscservice“, which corresponds to the service labeled “Offline Files” in the Windows services list.

The last step to get this policy working correctly is to add in a registry key that will fix machines that have already been used to cache certain network resources. Essentially adding this registry key tell the machine to blow up its database of offline files and tells the machine to remove the cached files as well. To configure this settings we need to add in a custom reg entry:

Computer Configuration -> Preferences -> Windows Settings -> Registry

Key: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\CSC\Parameters
Value name: FormatDatabase
Value type: DWORD
Value data: 1

Here is a good article with instructions on how change the registry settings by hand and a screenshot of my own GP environment with how the settings should look via the GP Management Console.

That should be all the necessary changes that need to be made. If I missed anything let me know, hopefully this will save people time in the future.