The purpose of my experiment was to evaluate the “entry” threshold when working with the JIRA product. So do not judge strictly when I say that I just decided to put jIRA in my infrastructure. Earlier I was engaged in setting up Jira to fulfill the needs of working with tasks, but I never installed it before. My knowledge of the installation of Jira boiled down to 3 points:

  • I knew that the software was cross-platform,
  • It’s written in Java
  • It should have worked with an external database.

I deliberately did not watch any instructions to evaluate how I could cope, relying only on an intuitive interface and 8 years of experience in administering servers and systems. Since I’m an expert in the administration of Windows systems, the choice of operating system for Jira Server was partly due to my laziness and cleanliness of the experiment, and partly by a desire to verify the validity of the fact that the JIRA in Windows consumes much more resources than its analog in Linux.

I understand that some of the over-scrupulous colleagues will say that reading readme.txt helps to prevent a lot of problems and saves time (for them I’ll nonetheless provide links to the installation instructions and in particular to the system requirements), but there are many who don’t do that. I think that the following information will be useful for various reasons. Well, let’s get started.

So, I downloaded the latest available version of JIRA (which was 7.3.1) on the prepared in advance Windows Server 2016 virtual machine (with 1 virtual processor, 1 GB of RAM and 30 GB of hard disk space). And then I launched the installation.

The system asked if I had enough experience to clearly answer it’s questions, and I felt that I could manage it.

On the next screens I have specified the path to the working file software and additional objects, such as backups, plug-ins, and attachments. I also chose the option of creating program shortcuts. I left the default port values unaltered and chose the option of installing JIRa as a Windows service. After thoroughly studying the screen with the contents of the selected settings before installation, I clicked the Install button.

After a couple of minutes, the installation has been completed and the system suggested to open the browser. The browser opened the page “Http://localhoost:8080”, and voila!

Jira does not open. I was starting to get upset and went to make some tea. Although updating the page with a hot cup in my hand had its magic: the page successfully loaded and suggested to continue the setup.

Since I wasn’t just an amateur, and I was planning to install and configure Jira and other Atlassian products in a productive environment, then my further choice was obvious.

My infrastructure already had a MySQL server, so my choice fell on it. By the way, it turned out to be the most interesting variant in terms of the complexity of installing Jira. It is required to download the additional driver solely for MySQL.

I clicked on the link that was shown on the screen, downloaded the suggested archive with the driver and unzipped the jar file to the LIB folder, located among the JIRA working files. Being not able to resist the interest, I looked in other paragraphs of the instruction for assembling MySQL for Jira nonetheless.

I checked the MySQL parameters:
and rebooted the system for the sake of good order. While the system was booting (in my case the start of the JIRA service took about 4 minutes and I watched this process by means of loading the page “http://localhost:8080/secure/SetupMode!Default.jspa”) I created a new database and user.

Note that the last screenshot is relevant for the case when Jira and MySQL are installed on the same host. In my case, the Jira server IP was specified in the Host field.

Do not forget that the case of letters matters. This fact has made me put some efforts into double-checking the correct settings before the test connection was successful.

By clicking Next, I moved the installation to the stage of creating the database. This process also turned out to be quite long, for the third time plunging me into thoughts about resources allocated for a virtual machine. And the error that was shown at the operation output made me take up this issue.

In fact, the “JIRA Startup Failed” error with the commentary “The following plugins are required by JIRA, but have not been started” can be caused by various problems. But in my case the 1 GB-sized memory was completely occupied. I rebooted the Atlassian Jira service. While it was restarting, another 1 GB of RAM was added to the virtual machine. The benefit of Hyper-V on Windows 2016 is that it can do this “on the fly.” The amount of memory added was chosen based on system requirements (which I had to read, yes).

Everything started to work. The system asked me to provide the name of the company and the URL for accessing jira. Here I would recommend using a domain name, preferably one that can be used both inside and outside the company. Even if you plan to use Jira only for the internal needs of the enterprise, it is likely that the scenario will require external access to this system.

At the next step, the issue at stake was the license type. You can use an existing license, or generate a trial license. The second option was suitable for the system and me. When it was activated, jira successfully took up the allocated system gigabytes of RAM, increased the CPU load to the maximum, and successfully opened the settings page of the administrative account. By the way, it decreased the CPU load, but the use of RAM was still at the level of 2GB.

Impatiently I added another 1 GB of RAM to the system. I set up an administrative account, for purposes of configuration of which the system requires an e-mail, which may well be a fake.

At the next step, I chose the option of configuring the notifications via e-mail later and clicked on the sacred Finish button.

Bingo! I finally see:
In this case, the load on the server does not display anything unusual:
In future articles, I will observe the resource consumption during the further configuration and use of Jira.

Conclusion. Installation of Jira is well within reach of an average technical expert. The load on the system during the installation period does not exceed the admissible values.

Of course, if you want to use it in a productive environment in the future, configure various modules and functionality, then it’s better to trust the installation and configuration to my skilled colleagues, [email protected]