Flatcar Container Linux recently announced the Stable support for ARM64.

Perfectly timed! I had a Raspberry Pi 4 lying around and had just ordered a few more to set up a home lab during the holidays. The newer Pis are yet to arrive, so better utilize the time writing a walkthrough on how to use Flatcar Container Linux on your Pis.

Hardware Requirements

• Goes without saying, a Raspberry Pi 4
• Form of storage, either USB and/or SD card. USB 3.0 drive recommended because of the much better performance for the price.
• Display (via micro HDMI/HDMI/Serial Cables), keyboard

⚠️ WARNING ⚠️ The UEFI firmware used in this guide is an UNOFFICIAL firmware. There is a possibility of damage caused due to the usage of this firmware. The author of this article would not be liable for any damage caused. Please follow this article at your own risk.

Update the EEPROM

The Raspberry PI 4 use an EEPROM to boot the system. Before proceeding ahead, it is recommended to update the EEPROM. Raspberry Pi OS automatically updates the bootloader on system boot. In case you are using Raspberry Pi OS already, then the bootloader may be already updated.

For manually updating the EEPROM, you can either use the Raspberry Pi Imager or the raspi-config. The former is the recommended method in the Raspberry Pi documentation.

We will also see later how the RPi4 UEFI firmware needs a recent version of EEPROM.

Using the Raspberry Pi Imager (Recommended)

• Install the Raspberry Pi Imager software. You can also look for the software in your distribution repository. Being a Fedora user I installed the software using dnf

dnf install rpi-imager

• Launch Raspberry Pi Imager.
• Select Misc utility images under Operating System.
• Select Bootloader.
• Select the boot-mode, SD, USB
• Select the appropriate storage, SD or USB
• Boot the Raspberry Pi with the new image and wait for at least 10 seconds.
• The green activity LED will blink with a steady pattern and the HDMI display will be green on success.
• Power off the Raspberry Pi and disconnect the storage.

Using the raspi-config

• Update the rpi-eeprom package.

sudo apt update
sudo apt full-upgrade
sudo apt install rpi-eeprom

• Run sudo raspi-config
• Select Advanced Options.
• Select Bootloader Version
• Select Latest for latest Stable Bootloader release.
• Reboot

Using the rpi-eeprom-update

• Update the rpi-eeprom package.

sudo apt update
sudo apt full-upgrade
sudo apt install rpi-eeprom

• Check if there are available updates.

sudo rpi-eeprom-update

• Install the update

# The update is pulled from the `default` release channel.
# The other available channels are: latest and beta
# You can update the channel by updating the value of
# `FIRMWARE_RELEASE_STATUS` in the `/etc/default/rpi-eeprom-update`
# file. This is useful usually in case when you want
# features yet to be made available on the default channel.

# Install the update
sudo rpi-eeprom-update -a

# A reboot is needed to apply the update
# To cancel the update, you can use: sudo rpi-eeprom-update -r
sudo reboot

Installing Flatcar

Install flatcar-install script

Flatcar provides a simple installer script that helps install Flatcar Container Linux on the target disk. The script is available on Github, and the first step would be to install the script in the host system.

mkdir -p ~/.local/bin
# You may also add `PATH` export to your shell profile, i.e bashrc, zshrc etc.
export PATH=$PATH:$HOME/.local/bin

curl -LO https://raw.githubusercontent.com/flatcar-linux/init/flatcar-master/bin/flatcar-install
chmod +x flatcar-install
mv flatcar-install ~/.local/bin

Install Flatcar on the target device

Now that we have the flatcar-install installed in our host machine. We would go ahead and install the Flatcar Container Linux image on the target device. The target device could be a USB or SD Card. In my case, I reused the existing SD Card which I used in the previous steps. You can use a separate storage device as well.

The options that we will be using with the scripts are:

# -d DEVICE   Install Flatcar Container Linux to the given device.
# -C CHANNEL  Release channel to use
# -B BOARD    Flatcar Container Linux Board to use
# -o OEM      OEM type to install (e.g. ami), using flatcar_production_<OEM>_image.bin.bz2
# -i IGNITION Insert an Ignition config to be executed on boot.

• The device would be the target device that you would like to use. You can use the lsblk command to find the appropriate disk. Here, I’m using /dev/sda which was in my case.
• With the given values of channel and board, the script would download the image, verify it with gpg, and then copy it bit for bit to disk.
• In our case, Flatcar does not yet ship Raspberry PI specific OEM images yet so the value will be an empty string ''.
• Pass the Ignition file, config.json in my case, to provision the Pi during boot.

{
  "ignition": {
    "config": {},
    "security": {
      "tls": {}
    },
    "timeouts": {},
    "version": "2.3.0"
  },
  "networkd": {},
  "passwd": {
    "users": [
      {
        "name": "core",
        "sshAuthorizedKeys": [
          <Insert your SSH Keys here>
        ]
      }
    ]
  },
  "storage": {
    "files": [
      {
        "filesystem": "OEM",
        "path": "/grub.cfg",
        "append": true,
        "contents": {
          "source": "data:,set%20linux_console%3D%22console%3DttyAMA0%2C115200n8%20console%3Dtty1%22%0Aset%20linux_append%3D%22flatcar.autologin%20usbcore.autosuspend%3D-1%22%0A",
          "verification": {}
        },
        "mode": 420
      }
    ],
    "filesystems": [
      {
        "mount": {
          "device": "/dev/disk/by-label/OEM",
          "format": "btrfs"
        },
        "name": "OEM"
      }
    ]
  },
  "systemd": {}
}

Write away!

sudo flatcar-install -d /dev/sda -C stable -B arm64-usr -o '' -i config.json

If you already have the image downloaded you can use the -f param to specify the path of the local image file.

sudo flatcar-install -d /dev/sda -C stable -B arm64-usr -o '' -i config.json -f flatcar_production_image.bin.bz2

Raspberry Pi 4 UEFI Firmware

rpi-uefi community ships a SBBR-compliant(UEFI+ACPI), ArmServerReady ARM64 firmware for Raspberry Pi 4. We would be using the same to UEFI boot Flatcar.

v1.17 of the pftf/RPi4 introduced two major changes:

• Firstly, it enabled firmware boot directly from the USB. This is particularly helpful if you are using the installation process using a USB device. To add a fun story, I dropped my Pi and broke the SD card slot. Until the Pi gets repaired, I’m making use of direct USB boot 😎
• Secondly, support for directly placing the Pi boot files into the EFI System Partition (ESP). This feature was not implemented in the firmware, rather from the upstream firmware from Raspberry Pi Foundation. This is why it is recommended to update the Pi EEPROM at the very beginning.

Let’s move ahead with the final steps.

• Place the UEFI firmware into the EFI System Partition.

efipartition=$(lsblk /dev/sda -oLABEL,PATH | awk '$1 == "EFI-SYSTEM" {print $2}')
mkdir /tmp/efipartition
sudo mount ${efipartition} /tmp/efipartition
pushd /tmp/efipartition
version=$(curl --silent "https://api.github.com/repos/pftf/RPi4/releases/latest" | jq -r .tag_name)
sudo curl -LO https://github.com/pftf/RPi4/releases/download/${version}/RPi4_UEFI_Firmware_${version}.zip
sudo unzip RPi4_UEFI_Firmware_${version}.zip
sudo rm RPi4_UEFI_Firmware_${version}.zip
popd
sudo umount /tmp/efipartition

• Remove the USB/SD from the host device and attach it into the Raspberry Pi 4 and boot.

Voilà! In no time, your Raspberry Pi would boot and present you with a Flatcar Container Linux prompt.

Further Reading

• rpi4-uefi.dev - RPi4 UEFI Firmware Official Website
• Raspberry Pi documentation

Photo by Anto Meneghini

I was talking to Jason today, discussing aspects of life and building habits.

Last week was pretty rough, devoid of motivation to maintain my habits. To which he said to me, “to fall in love with boredom” and passed me a couple of blog post from James Clear.

• How to Fall in Love With Boredom and Unlock Your Mental Toughness
• How to Stay Focused When You Get Bored Working Toward Your Goals
• Lessons on Success and Deliberate Practice from Mozart, Picasso, and Kobe Bryant

The first blog was one I could relate to in an instant. These days, the mind is easily, actually very easily distracted.

Everything looks good when you start, you have the sheer motivation in the beginning, or as in terms of relationship, the honeymoon phase. Once it fades, it’s possible the motivation fades away too. This is were the consistency plays a vital role.

James talks of a strategy known as “Seinfeld Strategy” based on Jerry Seinfeld. At the core Seinfeld Strategy talks about not breaking the chain, and marking on a calendar each single day to note the task has been done. Over the years of consistency, the mastery will follow.

Few examples.

• Paul Erdos, the fantastic mathematician, published over 1,500 papers before establishing himself as a thought leader.
• Famous composers put in 10 years of under appreciated work before earning recognition.
• Milo of Croton, the legendary Greek wrestler, picked up a young calf every day until he developed incredible strength.

Remember, it’s a constant fight between procrastination and motivation.

Photo by Priscilla Du Preez

Configuration Levels

• --local

  (Default) Local configuration values are stored and managed at the repository level. The values are stored in a file found in the repo’s .git directory: .git/config

  If you are planning to set some configuration only to be used by the specific repo you are in, then go ahead with using --local.

  git config --local user.email "sayan@mycompany.io"

• --global

  Global configuration is one level above Local. These configuration are user-specific and managed through the .gitconfig file in user’s home directory: ~/.gitconfig

  If you want your configuration to reflect to all your repo’s in the system, then go ahead with using --global

  git config --global user.email "sayan@emailprovider.io"

• --system

  System-level configuration applies to every user on the system, which means if none of the the local or the global values overrides the system level Git configuration value will be used. These configurations are true for all the repos’ in the system and is managed through the /etc/gitconfig file.

  git config --system user.email "infra@mywebsite.io"

Listing all the configurations

• To see all the configuration values, use the command:

  git config --list

• It can get quite messy when you have configuration values coming from multiple sources. To see the origin of the value as well add --show-origin

  git config --list --show-origin

Your Identity

The first chore the you always need to do after setting up Git is setting up your user name and email address. This is required because Git attaches this piece of information to ever changeset that you create in Git, known as Git commit.

Which level of configuration should we use? Because the details are user specific in most cases, the optimum answer is --global

git config --global user.name "Sayan Chowdhury"
git config --global user.name "sayan@emailprovider.io"

Let’s verify this in the global settings file:

$ cat ~/.gitconfig
[user]
    name = Sayan Chowdhury
	email = sayan@emailprovider.io

What are the other alternative will you use to verify?

Your Editor

Git invokes the editor for a lot of command for example while doing a commit, or rebasing your commits. When you type these commands it’s fire up the editor infront of you. Now, each of us has their own choice of editor so vim gives you option to configure which editor you would like to use.

I’m a vim user, thus for me the command would be:

git config --global core.editor "vim"

Colors in Git

Coming soon

Merge and diff tools

Coming soon

Git configurations you might find useful

More details on the following config coming soon

• alias
• commit.template
• core.excludesfile
• gpg.program
• help.autocorrect
• http.cookiefile
• push.current
• user.signingkey

–

To more on git config, read man git-config

Today, I found a cleaner solution to this problem. I can use insteadOf to enforce the connection via SSH.

git config --global url."git@github.com:".insteadOf "https://github.com/"

This creates an entry in your .gitconfig:

[url "git@github.com:"]
    insteadOf = https://github.com/

d46d2f8a (HEAD -> feature, devel/feature) Add Fedora system to handle the Fedora Cloud images
f726b033 vendor: Add new dep github.com/ulikunitz/xz
dbdbda67 Split release and pre-release to handle multiple systems

Here, d46d2f8a is the HEAD, followed by f726b033 and dbdbda67.

Your reviewers go through the commits, and suggest changes to some files. The suggestions now need to be applied to the commit dbdbda67. How would you fix this? This is a jam I usually land myself in, as for review I prefer my commits to be concise, shows incremental changes and for each commit to be atomic.

There are two ways you can use to fix this issue:

Rebase and Edit

I start off with git rebase -i which opens up the list of commits up for rebase. In our example here, the third commit from the HEAD needs to be changed, so I’ll rebase commits till HEAD~3.

git rebase -i HEAD~3

This opens up the editor, with the options

pick dbdbda67 Split release and pre-release to handle multiple systems
pick f726b033 vendor: Add new dep github.com/ulikunitz/xz
pick d46d2f8a Add Fedora system to handle the Fedora Cloud images

Below this, you will find a list of commands you can use with git-rebase:

# p, pick <commit> = use commit
# r, reword <commit> = use commit, but edit the commit message
# e, edit <commit> = use commit, but stop for amending
# s, squash <commit> = use commit, but meld into previous commit
# f, fixup <commit> = like "squash", but discard this commit's log message
# x, exec <command> = run command (the rest of the line) using shell
# d, drop <commit> = remove commit
# l, label <label> = label current HEAD with a name
# t, reset <label> = reset HEAD to a label
# m, merge [-C <commit> | -c <commit>] <label> [# <oneline>]
# .       create a merge commit using the original merge commit's
# .       message (or the oneline, if no original merge commit was
# .       specified). Use -c <commit> to reword the commit message.

The list is quite coherent, but for this post, we’ll use the edit command. So replace pick with edit against the commit you would like to edit and exit the editor.

So for our example, it’ll become (note the edit command in the first line):

edit dbdbda67 Split release and pre-release to handle multiple systems
pick f726b033 vendor: Add new dep github.com/ulikunitz/xz
pick d46d2f8a Add Fedora system to handle the Fedora Cloud images

# Rebase 51c8dc75..d46d2f8a onto 51c8dc75 (3 commands)
#
# Commands:
# p, pick <commit> = use commit
# r, reword <commit> = use commit, but edit the commit message
# e, edit <commit> = use commit, but stop for amending
# s, squash <commit> = use commit, but meld into previous commit
# f, fixup <commit> = like "squash", but discard this commit's log message
# x, exec <command> = run command (the rest of the line) using shell
# d, drop <commit> = remove commit
# l, label <label> = label current HEAD with a name
# t, reset <label> = reset HEAD to a label
# m, merge [-C <commit> | -c <commit>] <label> [# <oneline>]
# .       create a merge commit using the original merge commit's
# .       message (or the oneline, if no original merge commit was
# .       specified). Use -c <commit> to reword the commit message.
#
# These lines can be re-ordered; they are executed from top to bottom.
#
# If you remove a line here THAT COMMIT WILL BE LOST.
#
#       However, if you remove everything, the rebase will be aborted.
#
#
# Note that empty commits are commented out

What this will do is stop the rebase process at the commit you wanted to edit. At this moment, you can go ahead and perform your changes. Add files, modify them, remove them etc.

Once you’re done, add the files using git add and continue the rebase using:

git rebase --continue

The other way? Rebase and Fixup

I’ve recently started to use this method a whole lot more than the previous one I described. Here, you can just go ahead and do the changes.

Once done, pass the argument --fixup, while making the commit. For our example,

git commit --fixup=dbdbda67

This would create a fixup commit starting with !fixup. Next, go ahead and rebase the commits from the commit to be fixed to fixup commit.

git rebase -i --autosquash dbdbda67~1

This will merge the fixup commit with the commit to be fixed.

Voila! Serve the PR while hot. 🍜

Do you know easier methods to tackle this issue? Drop me a message on Twitter @yudocaa. I would like to thank Jason, Jaysinh, and Sayani who did proofreading for this blog post. Thanks to @anniespratt for the cover image.

2018, I started working on Golang, and vim had to adapt to my needs, but it did not quite succeed and I made my switch to neovim around the end of the year. Around the same time, I took a resolution to dive deep into neovim to increase my productivity.

Jump Lists

In golang, I heavily use a :GoDef which is part of the vim-go plugin to go to a specific symbol or declaration.

Vim keeps track of all the jumps (previously visited cursor positions). :jumps lists down all the performed jumps for the current window. Ctrl+O & Ctrl+I helps you to cycle through the jumps. But, what counts as a jump?

Any of the actions mentioned on the above list counts as a jump, and makes an entry into the jump list. You can clear the jump list using :clearjumps.

The columns being jump, line, column and file/text. Given the above:

• Ctrl-I to jump to line 415 in the current buffer.
• Ctrl-O to jump to line 358 in the current buffer.
• 3 then Ctrl-O to jump to line 364 in current buffer.
• 5 then Ctrl-I to jump to line 395 in the mantle/cmd/plume/prerelease.go.

I’ll keep on updating the posts as I learn more about vim/neovim. Till then, saraba da!

Reference for this post

• Jumping to previously visited locations
• Understanding Vim’s Jump List
• Cover Image by Ryan Pernofski

Gladly, the Fedora Kernel Team (Laura Abbott, Justin Forbes and Jeremy Cline) and Peter Robinson had already everything in place. The only missing piece was to add the support to fedimg to create arm64 based AMIs.

With the release of fedimg==2.4.0 (thanks to Patrick), the new AMIs were happily getting uploaded to AWS and with a slight delay, Fedora had the support for the arm64 along with the x86_64 AMIs.

Known issues?

One of the issues which I faced was the instance getting ready takes some time which is almost 5 minutes. But work is on the way to make the experience better.

Availability?

A1 instances are currently supported only in selected regions: US East (N. Virginia), US East (Ohio), US West (Oregon), Europe (Ireland).

Also not all the availability zones are supported, so while launching instances you might see this error:

An error occurred (Unsupported) when calling the RunInstances operation: Your requested instance type (a1.large) is not supported in your requested Availability Zone (us-west-2b). Please retry your request by not specifying an Availability Zone or choosing us-west-2c, us-west-2a.

which I found linked to the subnet’s availability zone. So, you might need to change the subnet’s availability zone to launch the instance.

For any more issues, drop into #fedora-arm or #fedora-cloud IRC channel on Freenode.

Read the official announcement here.

The year I graduated from a third-tier college in Durgapur. I did not expect much from the college, mostly because of the poor placement record in the college.

To get myself a job, I was primarily going through hasjob.co and any job opportunities that came through Bangpypers mailing list.

Finally, I landed myself into an internship and later which turned into a full-time at HackerEarth was through the hasjob.co.

2018

5 years later, finding a job in a suitable company is still tough. In the last 5 years, I have come to know about more options, be it AngelList, Hasjob, StackOverflow, Facebook Job Groups etc.

But the problem still I think is the same, even if you apply through these websites a lot of times you don’t get a reply from the company you applied to. On the other hand, referrals work. I’ve referred good people to companies I know. This way, there is a chance of getting an interview in the company. Cracking the interview is a different story altogether which I believe depends a lot on the candidate.

Recently, I got to know that Prashant, one of my school senior and a close friend, who famously known in the tech community for his “Bitcoin Wedding” started an effort called “SkipTheLine”.

SkipTheLine is a newsletter where Prashant publishes profiles of three developers. These developers are from the community, who are active in open source, or competitive programming, or just good at technologies they work on. He goes on to introduce the developers through email with the companies point of contact via email and then the candidate and the POC take the discussion forward.

I personally loved the initiative that he took as at the end of the day if people whom I know come asking for a job referral I would just direct them to SkipTheLine. Prashant has quite a stronghold in the startup community and does great work in connecting the folks with some really good startup across the country.

I know people personally who got hired a within few days of their newsletter published so if you are looking out for a job, do fill out the SkipTheLine form.

If you are looking to hire, do drop me an email at gmail AT yudocaa DOT in.

With the Fedora 28 release, the Fedora AMIs come with the ENA support.

To check whether an AMI has the Enhanced Networking with ENA-Enabled, you can query the AWS EC2 API, using the AWS Command Line Interface. You need to check if the enaSupport attribute set. If the attribute is set, the response is true.

➜  ~ aws ec2 describe-images --region us-east-1 --image-id ami-cceb7ab3 --query 'Images[].EnaSupport'
----------------
|DescribeImages|
+--------------+
|  True        |
+--------------+

Also with Fedora 28 release, Fedora deprecates the release of paravirtual (PV) images.

What’s the problem?

fedimg boots an utility instance using the RHEL AMI configured for that region.

The problem with this design is that while adding a new region we need to look for the appropriate RHEL AMI in the newer region first. Afterward, we need to test if fedimg works with the newer regions RHEL AMIs. Finally, if everything turns out to be okay we can go ahead with adding the AMI to the fedimg configuration.

This becomes a tedious task over time and delays adding a new region.

How does redesign fixes this?

The new design completely removes the dependency of the utility instances. fedimg would then utilize euca-import-volume to upload the raw cloud image to create a volume via S3.

fedimg will keep retrying to check if the volume is read using euca-describe-conversion-tasks. Once the volume is created, fedimg goes ahead with creating the snapshot of the volume and then finally the AMI.

This change results in removing a lot of redundant code and a simpler fedimg configuration file [1][2]. I have been working on this for sometime now.

Adding a new region now becomes effortless as we just need to append the region to the configuration file.

Fedimg also right now boots up test instances to test the AMIs and does a basic /bin/true test. This test just ensures that the machine boots and nothing more than that.

In the next post, I will be writing on how I will be going ahead to build a testing infrastructure for the AMIs using Autocloud and ResultsDB.

Fedimg is a Python-powered service that uploads the Fedora Cloud images on various Cloud providers. such as Amazon Web Services (AWS) .

How does it currently work?

Fedimg listens to fedmsg for nightly as well as the production compose builds messages. On receiving a new message, the FedimgConsumer triggers the upload process. fedimg uses fedfind to extract the metadata for the compose from where it get’s the URL for the compressed raw image files.

The AWS EC2Service then starts the process of creating the AMI.

• Booting up the utility instance.
  The EC2Service parses the fedimg configuration file to get the list of configuration regions mapped with the RHEL AMIs, AKIs, architecture etc. EC2Service.upload starts the uploading by booting a utility instance in the first region(determined from the config file). The process continuously keeps trying to ssh into the machine to check if the machine has boot-ed up.

• Downloading image & taking snapshot
  Once the utility instance is ready to use the compressed RAW image is downloaded in the secondary volume. A snapshot is taken out of the volume and the AMI is created out of it.

• The AMIs are then copied to the other regions

• Testing the AMIs Once the AMIs are created, a very basic /bin/true test is performed on a test instance created using one of the created AMIs. Once the tests passes, the AMIs are made public.

At any point doing the execution, if the code throws an exception and even if the AMIs fails the _clean_up method is fired. The method cleans up the utility instance, test instance and the volumes which were created.

The next blog post will discuss the issues related to the current architecture and about the new architecture.

Previously, it used to listen to fedmsg for successful Koji builds. Whenever, there is a new message the AutocloudConsumer queues these message for processing. The Autocloud job service then listens to the queue, downloads the images and runs the tests using Tunir. A more detailed post about it’s release can be read here.

During the Fedora 24 cycle things changed. There was a change on how the Fedora composes are built. Thanks to adamw for writing a detailed blogpost on what, why and how things changed.

With this change now autocloud listens to the compose builds over the fedmsg. The topic being “org.fedoraproject.prod.pungi.compose.status.change”. It checks for the messages with the status FINISHED and FINISHED_INCOMPLETE.

After the filtration, it gets the Cloud Images built during that particular compose using a tool called fedfind. The job here is parse the metadata of the compose and getting the Cloud Images. These images are then queued into both libvirt and vbox boxes. The Autocloud job service then downloads the images and run the tests using Tunir.

Changes in the Autocloud fedmsg messages.

Earlier the messages with the following topics were sent

• autolcoud.image.queued - When the images are queued for testing
• autocloud.image.running - When the tests for the images are running
• autocloud.image.success - When the tests for the images passes.
• autocloud.image.failed - When the tests for the images failed

Now along with the fedmsg message for the status of the image test. Autocloud also sends messages for the status of a particular compose.

• autocloud.compose.queued - When the composes is in queued state
• autocloud.compose.running - When the images for the composes are in testing phase
• autocloud.compose.complete - When running the test for the images for a compose finishes and contains the aggregate result for the tests

The compose_id field was added to the autocloud.image.* messages

Changes in the UI

• A page was added to list all the composes. It gives an overview of the composes like if it’s still running, number of tests passed, etc
• The jobs page lists all the tests data as earlier. We added filtering to the page so filter the jobs based on various params
• You need to agree that the jobs output page looks better than before. Now, rather the showing a big dump of text the output now is properly formatted. You can now reference each line separately.

Right now, we are planning to work on testing the images uploaded via fedimg in Autocloud. If the project looks interesting and you are planning to contribute? Ping us on #fedora-apps on Freenode.

Weechat setup turned out to be super simple and out of the box. I am running Weechat within a tmux session. But the problem I was facing was due to poor internet connection (thanks to Airtel for this and this was my primary reason for switching to weechat from Xchat) I got disconnected from the DO box quite often.

So, I looked over for a solution to get IRC notfications to my mobile device and Weechat seems to already have a script for that called pushover.pl.

So, The script pushes the notifications from weechat to your device via Pushover.

Installing the script

1/script install pushover.pl

Create an account in Pushover, verify your email and create an application. Once the application is created you find your application token, and your user key.

Configuring

1/set plugins.var.perl.pushover.token <your application token>
2/set plugins.var.perl.pushover.user <your user key>

Now, your weechat is set to send messages through Pushover, install the application on your mobile device and you will start receiving the irc messages whenever you are mentioned or a private message.

The script also offers other customizations like

 1'enabled' => ['on', "Turn script on or off"]
 2'service' => ['pushover', 'Notification service to use.']
 3'token' => ['', 'pushover API token/key']
 4'user' => ['', "pushover user key"]
 5'nma_apikey' => ['', "nma API key"]
 6'pb_apikey' => ['', "Pushbullet API key"]
 7'pb_device_iden' => ['', "Device Iden of pushbullet device"]
 8'sound' => ['', "Sound (empty for default)"]
 9'priority' => ['', "priority (empty for default)"]
10'show_highlights' => ['on', 'Notify on highlights']
11'show_priv_msg' => ['on', 'Notify on private messages']
12'redact_priv_msg' => ['off', 'Hide the private message text']
13'only_if_away' => ['off', 'Notify only if away status is active']
14'only_if_inactive' => ['off', 'Notify only if buffer is not active']
15'blacklist' => ['', 'Comma separated blacklist for notifications']
16'verbose' => ['1', 'Verbosity level']
17'rate_limit' => ['0', 'Rate limit in seconds']
18'short_name' => ['off', 'Use short buffer name in notification']

Fedimg listens for the messages from Fedmsg for the Koji image builds.

Plan of work

• Add scripts to remove old AMIs we have uploaded (https://github.com/fedora-infra/fedimg/issues/37)

  Storing the AMIs are costing us lot. So the Cloud WG has decided to delete the older AMIs. The deletion criteria would be

  • Delete the pre-release AMIs after the final release
  • Delete the nightly build AMIs after 5 days of build

• Reorganize the way we upload images to AWS (https://github.com/fedora-infra/fedimg/issues/42)

  Currently fedimg boots up an instance, and downloads the image to volume and then takes the snapshot to create the AMI.

  gholms suggested that instead we can use euca2ools and using the command euca-import-volume we can directly create a volume out of the image. Then take the snapshot and create the AMI.

• Add generic re-try logic to all of fedimg’s operations (https://github.com/fedora-infra/fedimg/issues/34)

  Currently, the upload method for the EC2 Service is a big chunk of code. My plan is to break code like these into different chunks and add a retry decorator to it. The decorator will handle things like delay, number of times to run etc.

• Get autocloud to do the testing on AMIs (https://github.com/fedora-infra/fedimg/issues/43)

  Currently, Tunir has the feature to test the AMIs but it’s in the development branch and before starting off with this issue we need to work on autocloud too in between to integrate the feature to test AMIs.

• Generate a work report at the end of a job (https://github.com/fedora-infra/fedimg/issues/26)

  This is an easy-fix to add to add standard logging and can be completed with the other tasks in progress.

• Add secondary cloud providers (https://github.com/fedora-infra/fedimg/issues/10)

  This issue is blocked on legal stuff. So, as I work on other issues I will be checking the status for this ticket and can swap some low-priority issue with this and write up the service which is clear to go.

Interested?

If you are interested to help out you can ping on IRC in #fedora-apps on Freenode.

Autocloud publishes the messages to fedmsg when the image is queued, running, success and failed.

What is Bugyou?

Bugyou files issues for failed tests from Autocloud. Bugyou is part of Fedora Project. It listens to Autocloud messages from fedmsg and files bug for images which have failed the tests. Bugyou keeps updating the status of the images in the bug and closes the issue once the tests are successful.

Currently Bugyou files issues in the repo atomic-images in Pagure

If you want to learn about how it is deployed, please visit this page

Future plans

The plan is to make Bugyou more generic and to have plugins (Autocloud, Koji) and various issue tracking tools as services like Pagure, Github, Trac etc.

I have been contributing to the Fedora for couple of years now starting from the contributing to askbot-devel while we were gearing up for the Ask Fedora release. This year in July, I along with kushal, rtnpro, praveenkumar started working on Autocloud, as a part of the Two-Week Atomic during the Fedora 23 cycle. So, I last couple of years of I have met and worked with a lot of people in the Fedora community. Joining Fedora was like working with friends.

I joined as a remotee working from Pune. I will be primarily working on different applications in the Fedora Infrastructure

First Week

I was handed over a Lenovo Thinkpad T450s, it turned out to be a good working machine with good configs. I started using mutt as my email client, pass for managing passwords, taskwarrior and bugwarrior to manage my tasks and bugs.

Chalked out a plan for Bugyou and worked on a POC. Bugyou is an automatic bug reporting tool that consumes the messages from fedmsg and file bugs to different bug tracking tools for registered topics. I wrote down a fedmsg consumer that would be consuming the failed and the successful messages and file tickets from them pagure.

Second and Third Week.

During these two weeks, my primary focus was on deploying Bugyou. I fixed a couple of bugs in libpagure, which is a Python library for the Pagure APIs. I started packaging python-libpagure and making it ready for deployment as it was dependecies for Bugyou. python-libpagure being my first RPM package I spent a lot of time reading through the guidelines, fixing the issues in accordance to the review given in the review request.

I added some management commands for Autocloud to abort long running task and also to re-queue tasks. I also started with learning golang and wrote a minimalist logging framework from brume.

We also had a Python3 Porting FAD on November 14-15 where I ported the python-retask to Python 3. Retask is a python module to create and manage distributed task queue/job queue.

Fourth Week

I worked with a feature in pagure to show the list of commits before a particular commit hex. I gave a demo of Bugyou the Fedora Release Engineering Tools Demo. At the end of the week, I had my New Hire Orientation programme where I learnt more about the culture, people and met new people at Red Hat.

It seems to be exciting road ahead!.

Currently, I am working on a new architecture for Bugyou. If you are looking forward to help, come to #fedora-apps on freenode.

2013 and internship

My college days were coming to an end with placements all around. I was sure to join in a startup. One fine day, I saw a job posting on hasjob on 12th December 2012 that boldly said “HackerEarth is buidling its initial team - Python/Django enthusiast needed”. The idea made me apply to HackerEarth and after a few rounds of email with Sachin and Vivek. I landed up in a remote intern position.

My first task at HackerEarth as an intern was to implement Auth/Profile system for Codetable. Damn! My first task itself was a daunting task. There was so many permutaiton of things to take care of private digest, public digest, read-only mode, read-write mode, authenticated user, anonymous user. 10 days into the internship the auth system was finally working. Finally, the code was deployed and the feeling was amazing.

Four months into the internship, I got a full time position at HackerEarth. I joined the team at their Bangalore office in July 2013.

Working at HackerEarth

The office I joined was Bangalore Alpha Labs, Koramangala. In the span of two years, I worked on a couple of things. I have worked on the Recruiter Panel, Plagiarism Detection Engine, Profile pages, Challenges pages, HackerEarth Shorterner and for the past one year I have been working on HackerEarth Sprints.

HackerEarth has scaled at a very rapid rate and the engineering team at HackerEarth to learn at a more faster pace to keep things going. At HackerEarth, as a full-stack engineer I got to know the deep knowledge of what it takes to keep a website running at such a huge scale.

Life at HackerEarth

Supercalifragilisticexpialidocious! I never felt here that I was out of college. Food fests, hardware hacking, TT, packs of Redbulls, movie and gaming nights and flexible timings. Anything that had to keep a developer going we had it all.

What’s next?

I have been using Fedora since the initial days of my college. An year, I was contributing to Fedora. I have been working on Fedora Infrastructure projects after regular office hours and weekends. I have been also involved in evangelizing Fedora and organizing Fedora meetups.

But now I will be joining the Fedora Engineering Team working on Fedora as a part of my day job. Looking forward to scale more mountains and see what rocks life throws at me at my new role at Fedora.

reptyr is a utility for taking an existing running program and attaching it to a new terminal. Started a long-running process over ssh, but have to leave and don’t want to interrupt it? Just start a screen, use reptyr to grab it, and then kill the ssh session and head on home.

The package available in Fedora/Ubuntu repositories.

% sudo yum install -y reptyr        # For Fedora users
% sudo apt-get install -y reptyr    # For Ubuntu users

The steps to migrate a process is

• Sending the current foreground job to the background using CTRL-Z.
• List all the background jobs using jobs -l. This will get you the PID

% jobs -l
[1]  + 16189 suspended  vim foobar.rst

Here the PID is 16189

• Start a new tmux or screen session. I will be using tmux

% tmux

• Reattach the background process using

% reptyr 16189

If this error appears

Unable to attach to pid 16189: Operation not permitted
The kernel denied permission while attaching

Then type in the following command as root.

% echo 0 > /proc/sys/kernel/yama/ptrace_scope

These are commands are compatible with screen also

For those who are new to Waartaa, Waartaa is an open-source communication tool for teams and groups. It is built on top of IRC. Currently, Waartaa is an IRC client as a service and it supports centralized logging, 24x7 idling, notifications and unique identity to a user on IRC. The source code is available at https://github.com/waartaa/waartaa for you to download and hack on it.

If you find Waartaa interesting and something that you need, please feel free to ping us at #waartaa on Freenode. If you are a student, you are most welcome to submit your project idea for GSoC 2014 to help make Waartaa awesome.

Mission

Our mission is to implement a one-stop open-source communication and collaboration tool for teams and groups, built on top of open source technologies (IRC, WebRTC, etc.).

Current Status

Waartaa is no longer just an idea in pages, it’s real, it’s live. You can try our demo instance at https://try.waartaa.com. We have also started to receive quite some contributions as well. Lately, Waartaa has seen some massive improvements in terms of optimization and scaling. You can download and run Waartaa in your server or machine very easily. This allows you to get your own IRC client as a service up and running in no time.

Technology Stack

• Meteor JS http://www.meteor.com/
• MongoDB http://www.mongodb.org/
• Forked node-irc https://github.com/waartaa/node-irc
• And a host of meteorite apps from https://atmosphere.meteor.com/

Roadmap

• Build a central hub for searching/reading channel logs for Open Source communities and projects.
• Build a faster and scalable backend.
• Freedom of choice: Expose an API so that users can use their existing IRC clients with waartaa.
• Find a secure way to authenticate with IRC services without storing RAW passwords.
• Respect user privacy: user personal messages should be stored in an encrypted format in the server.
• Allow users to download chat logs in various formats compatible with popular IRC clients.
• HTML5 mobile app
• VCS, Bugzilla and other task management tools integration.
• Video/audio conference facility on top of HTML5 and JS technologies.

Communicate

• IRC: #waartaa on Freenode
• Report bugs or submit patches:
• https://github.com/waartaa/waartaa/issues
• Mailing list: waartaa@googlegroups.com

This is a re-post of https://www.waartaa.com/blog/waartaa-applies-for-gsoc-2014.html

Where are the Kolkata Bloggers? Like, seriously ?

In August 8, 2009, Webreps organised Kolkata Bloggers Meet. That time I joined Webreps. In October 31, 2009, we organised the BCET Bloggers workshop in BCET. There BCREC and NITDgp also joined in. Have been active ever since then.

In 2014 - I hardly find any blogger visible and audible. I want to organise a Kolkata Bloggers Meet later this year. Any help shall be welcome. Please connect to this page: https://www.facebook.com/pages/Kolkata-Bloggers/401101290026684

I dont know why I do not feel good about this scenario. Kolkata Bloggers make a sad story. There might be many bloggers. We have incredibly good writers. But the BUZZ is so not there. There is no enthusiasm. There is hardly any blogging for the sake of blogging, the fun or the enthusiasm for it. Let’s find them from where ever they are hidden in Kolkata. Let’s meet up.

Check this screenshot of indiblogger, how big is Kolkata?

This screenshot makes me really curious to know on what happened to the Kolkata bloggers. So here’s an open letter to bloggers from Kolkata.

Dear Kolkata Blogger,

“Kolkata Bloggers” is an initiative to collect and unite Kolkata based bloggers. Our first aim is to connect all bloggers. This is why we are sending this open letter to you. As of now, we do not have a website. We have a page. And I would tell you how that might help you and the entire newer generation from Kolkata.

This initiative will help you reach more audience: If you are a new blogger, this page can help you get exposure. We are adding the RSS feeds of your blog to the Facebook page. Each time you update your blog, your blog post is automatically shared by our page and you reach to more number of viewer. If you are an existing blogger, you can guide us spread the awareness among the youth of Kolkata. You can always reach to newer audience in this way.

This page shall be motivating for young writers who do not blog. It is time we spread the awareness and help groom newer bloggers from the city. I often see good writers not taking the plunge to blog, but they have a wish to do it. This initiative, if properly supported, I believe will motivate them go to the next step.

This initiative might help you network better. We aim to build this as a networking platform for bloggers from Kolkata, both online and offline. I often see fantastic writers who are bloggers not connected with the rest of the community. Networking is fun and you reach to more people.

Be inspired: In grand bloggers meet, we aim to bring the masters of this field in front to hold seminars, talks, ideating sessions and workshops. This will help you 0to better your blogging skills.

Help companies organise Blogger engagement programmes in Kolkata: When it comes to Blogger engagement programs and brand promotions, Delhi Mumbai Pune and Bangalore are often preferred over Kolkata.

Learn. Network better. Reach more audience. Build the ecosystem.

How to join Kolkata Bloggers?

This is the link to the page: facebook.com/KolkataBloggers

Message us your Name, Blog URL, Twitter URL, Facebook URL, Google+ URL, email address (very required), Phone number. There is “Kolkata Bloggers badge” in the page as a pinned post. Put the badge on your blog. Let us know once you are done. We will add the RSS feed of your blog to our page. Stay connected. If you require any help, drop a message to the page. We’ll get back to you.

Best Regards,
Kolkata Bloggers team.

After getting through Google Summer of Code and getting covered in Super Student, a lot of people contacted me. All of them had the same set of questions.

1. How to get started?
2. I know x,y,z languages. which project should i contribute?
3. How can i filter out a bug?

So, I thought of writing a blog post in which i’ll try to answer the questions. But, first let us overview on what is Open Source Software.

What is Open Source Software(OSS)?

Open Source Software is a computer software whose source code is made publically available for modification and enhancements released under various licenses. “What is open source?” is a good read.

I believe in the fact that reading few lines of code is more worth than read 500 pages of a book. When you follow code written by great programmers you automatically tend to pickup some great practices while coding.

Even before I studied the subject Software Engineering in college, I had practical knowledge of version control systems, debuggers, issue trackers, continuous integration tools and this was all because of contributing to a open source project.

You make a great resume that stands out from the crowd. Contributing to open source software helps you building online presence.

But, these merry words really are not much help to get you started.

How do I get started?

The first and the foremost thing is choose a programming language of your choice. Once you are done with selecting a programming language, search for a project you might be interested.

Open Hatch is just for beginners like you. The search page of Open Hatch filters out bugs based on

• Languages
• Projects

Mozilla too has a wide range of projects filter based on programming language. What can I do for Mozilla?

How do I filter out a bug?

Truly speaking, finding a bug for a beginner is really a hard job. I faced the same problems. But i’ll be listing down on how to find easy bugs and get started with contributing a large organization.

Fedora

Fedora Easy Fix page list down all the easy fixes you can get started with. Each project lists down whom you need to contact and the maintainer of the project.

Fedora Infrastructure respositories contain some easy fixes. Look for the EasyFix label in the repository issues section. https://github.com/fedora-infra

Mozilla

Getting started with Mozilla is damn simple. Mozilla has multiple gateway for their fellow new contributors.

1. What can I do for mozilla?

This site suggests you project based on the programming skill you select. Once you decide you project it takes you to the respective mozilla project.

2. Bugs Ahoy!

Bugs Ahoy is a site particularly for the new contributors. The website categorizes the ‘Easy bugs’ and ‘Mentored Bugs’ based on the various project like JS Engine, Devtools, Firefox OS etc.

But, what if you don’t know which project is based out of which programming language. Bugs Ahoy categorizes the easy bugs based on the programming languages too primarily Python, Java, Shell, JS, C/C++, HTML/CSS. One can combine the combination of filters to choose the perfect bug to get started.

3. Mozilla Getting Involved Pages.

Mozilla has Getting Involved which list down the projects in a particular team The projects contains

• The mentors who can be contacted
• Bugzilla page
• IRC channel name
• List of mentored bugs
• Getting Involved DevTools - https://wiki.mozilla.org/DevTools/GetInvolved
• Getting Involved Mozilla.org - https://wiki.mozilla.org/Webdev/GetInvolved/mozilla.org
• Getting Involved AMO - https://wiki.mozilla.org/Webdev/GetInvolved/addons.mozilla.org
• Getting Involved SUMO - https://wiki.mozilla.org/Webdev/GetInvolved/support.mozilla.org
• Getting Involved MDN - https://wiki.mozilla.org/Webdev/GetInvolved/developer.mozilla.org
• Getting Involved Socorro - https://wiki.mozilla.org/Webdev/GetInvolved/crash-stats.mozilla.org
• Getting Involved Mozillians - https://wiki.mozilla.org/Webdev/GetInvolved/mozillians.org
• Getting Involved ReMo - https://wiki.mozilla.org/Webdev/GetInvolved/reps.mozilla.org
• Getting Involved input.mozilla.org - https://wiki.mozilla.org/Webdev/GetInvolved/reps.mozilla.org
• Getting Involved careers.mozilla.org - https://wiki.mozilla.org/Webdev/GetInvolved/careers.mozilla.org
• Getting Involved QMO - https://quality.mozilla.org/docs/webqa/get-involved/

Here is another link for Bugzilla Mentored Bugs list - https://bugzil.la/sw:mentor

KDE

If you want to start contributing to KDE, look out for the Junior Jobs tag. Like each organization KDE too contains a section on how to get started with the KDE projects.

• Getting Started - http://techbase.kde.org/Contribute
• Building a KDE Application - http://techbase.kde.org/Getting_Started/Build
• KDE Guide for a new contributor - http://flossmanuals.net/kde-guide/
• IRC - #kde-devel on Freenode
• Mailing List - https://mail.kde.org/mailman/listinfo/kde-devel

OpenStack

OpenStack is also a great project to start with. The OpenStack project is divided into various components: Swift, Glance, Nova, Horizon, Keystone etc. Each of these components have their own page. If you head over to the OpenStack Wiki Main page, you can see the components listed seperately.

Each Component page contains the link to repository, bug tracker(Launchpad), Documentation etc. OpenStack call their beginner bugs as “Low Hanging Fruit”.

• Getting Started - https://wiki.openstack.org/wiki/How_To_Contribute
• If you are a developer - https://wiki.openstack.org/wiki/How_To_Contribute#If_you.27re_a_developer
• Documentation - docs.openstack.org
• IRC - #openstack-101, #openstack on Freenode

And to find the low hanging fruits? Click here!

Wikipedia

Most of the people on the planet have heard of the name Wikipedia but ever thought of contributing to it. Wikipedia is supported by Wikimedia Foundation. Mediawiki tag their easy bugs with the easy tag. Click!

• Getting Started - http://www.mediawiki.org/wiki/Gerrit/Getting_started
• How to become a Mediawiki Hacker - http://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker
• Installing Mediawiki on local system - http://www.mediawiki.org/wiki/MediaWiki-Vagrant
• IRC - #mediawiki

Now this link is important

• Annoying little bugs - http://www.mediawiki.org/wiki/Annoying_little_bugs

GNOME

‘Gnome-love’ is the loving tag for the easy bugs who want to get started with the GNOME. If you ever visit the GNOME homepage: the “Getting Involved” link is clearly visible in header menubar. Anyways here is the link if you don’t want to search :)

GNOME maintains a guide for the newcomers: https://wiki.gnome.org/NewcomersTutorial

If you want to know on how to submit your first patch, read this : https://wiki.gnome.org/GnomeLove/SubmittingPatches

If you want to build the code, then just build it from here: https://wiki.gnome.org/GnomeLove/JhbuildIntroduction

Documentation: https://developer.gnome.org/

Apache

Apache maintains quite a number of projects. project.apache.org lists a catalog of more than 140 projects in which one can start contributing. The New Comers section is an easy read on how to get started in Apache, http://community.apache.org/newcomers/index.html

Getting Involved wit Apache Software Foundation: http://www.apache.org/foundation/getinvolved.html

I am an active Python/Django developer for a couple of years. So, I’m adding on how can one start contributing to Python / Django

Python

To start with contributing with Python, just head over to the Developer’s Guide (http://docs.python.org/devguide/ ). It starts with cloning the repository in your system tell goes on to tell how can you check for the easy fixes.

The developer guide mentions that how can you start with making documentation fixes, then move up to fix small bugs. Once you get acquainted you can start taking up easy bugs.

• READ: http://docs.python.org/devguide/fixingissues.html
• Issue Tracker: http://bugs.python.org/
• Easy Issues: http://goo.gl/NzJuDp

Django

Django is a Python based web-framework used by some really big companies e.g. Disqus, Instagram, Transifex etc. To start contributing in Django, head over to the page Contributing to Django.

This page lists down the links to mailing list, IRC channel, Issue Tracker.

• Getting Involved in Django: https://code.djangoproject.com/#Gettinginvolved

Like most of the open source project Django too has easy fixes for the new contributors.

• Little Easy Improvement - https://code.djangoproject.com/wiki/LittleEasyImprovements
• Easy pickings - https://code.djangoproject.com/query?status=!closed&easy=1

The “Little Easy Improvement” also mentions issue for which one can send patches for documentation, writing tests, or improving the codebase.

Django maintains a page to advice new contributors: https://docs.djangoproject.com/en/1.6/internals/contributing/new-contributors/

Contributing to Open Source does not enforce you to contribute only to a well-known community. There is a huge collection of open source projects available on websites like Github, SourceForge , Google Code, Bitbucket etc. I also enforce on the fact that just open source contribution does not help. One need to work on side-project for ideas to keep flowing out of his brain.

I also you like to quote a few lines from a nice blog post

You shouldn't become an artist so you can be famous, but because there's art inside of you that will kill you if you don't let it out. You shouldn't found a startup to make money, but because it's your life's work. And you shouldn't hack on open source projects because someone told you that your GitHub profile is your new resume, but because you want to code socially.

Once you start contributing, you start interacting to people via IRC, mailing lists. I strongly suggest you to go through the etiquette an organization follows. In general roll your eyes through this presentation to get a brief of it. http://www.shakthimaan.com/downloads/glv/presentations/mailing-list-etiquette.pdf

Even after reading this if you face any problem. Feel free to reach me on twitter or shoot me a mail at sayan DOT chowdhury2012 AT gmail DOT com. I would also like to thank Chandan Kumar who helped me in writing this blog post, he can be reached on chandankumar DOT 093047 AT gmail DOT com.

Lot of things to do!

Lot of things to do!

This was continuously revovlving around my head for last 2 weeks. I had to deploy the work i had done on staging instance and test it. But wait! I had never written a setup.py before, never made a RPM package before.

While in college, I had a fascination toward RPM packaging as most of my college seniors had created a RPM package sometime or other. During Mukti 2011, I had attended a RPM packaging workshop by Rahul Sundaram(mether), so i had the idea of how to package a RPM.

But well before that i had to write a setup script for darkserver. Though the basic setup.py did not take much time, but i got stuck when it came to include staticfiles in setup.py. After a lot of googling and seraching, the result was null. So i asked in #python IRC channel, when Ned Batchelder helped me out. Finally, I got a link to package a Django app titled- Packaging a Django resuable app

After completing writing setup.py, I started working on building RPM packaging. Here are some links:

• Packaging:Guidlines
• How to create an RPM package
• Packaging:Python

After completing rpm packaging, it’s time for koji scratch build. And i did it but while installing it in the staging instance, something happened. I had used target as rawhide instead of dist-6E-epel while koji scratch build. Nevermind, I repeated the steps and voilà, i successfully perform a fresh install of the package and everything was working properly.:)

And, with that Google Summer of Code came to an End.

Darkserver’s command line dashboard now supports the secondary architecture. You can easily shutdown the darkproducer of specifying which architecture to shutdown.

dark> shutdown darkproducer arch

The support for multiple buildqueues was removed.

After a rigorous testing on my laptop, i pushed the code to the dev instance that was assigned to me. But, a mysterious error happened, nor the pid or the daemon of darkproducer was getting created. Due to which the complete project came to a halt. Later, it was figured out that i was passing the config (--config) argument as a relative path of the conf file, rather than the absolute path which created all this trouble. So, the proper way to start the darkproducer is:

$ python darkproducer --start=157435 --config=/home/sayan/Development/darkserver/configs/darkserverurl-arm.conf

Darkserver also now has a web dashboard which shows the status of the darkproducer, darkbuildqueue and the darkjobworkers.

Method Not Allowed This is an XML-RPC server. Only POST requests are
accepted.

The project required a heavy amount of download, so my mentor gave me a dev instance to test out my code, all seemed to be fine, with the code properly working on local machine. But on the dev instance a major bug surfaced, all the architecture with referring to the same key. So, currently i started working on implementing different darkproducer-ids for different darkproducers.Also, I need to get started with a long-term bug, my mentor pointed me in the shutdown functionality of darkserver.

After recovering, i was back to coding/resolving issue in darkserver. Starting from the mid of week three, I started working on implementing the support for secondary architecture. Darkserver now supports secondary architecture(arm, ppc).

Defining the current structure of Darkserver: | darkproducer

$ darkproducer --start=BUILD_ID --config=CONFIG
Options:
-s BUILD_ID, --start=BUILD_ID Specifies the build id
-c CONFIG, --config=CONFIG Specifies the Config file

The config files points to either of the configuration files for Koji, ARM, PowerPC. The files being darkserverurl-koji.conf, darkserverurl-arm.conf, darkserverurl-ppc.conf placed under configs/. The default being for koji.

A ‘url’ key was added to retask Queue to maintain the relation between the build_id and the architecture.

The command line arguments in Darkserver are currently now parsed by optparse python module. So, a pid is added to darkbuildqueue and darkjobworker, so the command line arguments now is:

darkbuildqueue

$ darkbuildqueue --start --pid=1
Options:
-s, --start start the buildqueue
-p PID, --pid=PID specify the pid for build queue

darkjobworker

$ darkjobworkder --start --pid=1 // To start job worker 1
$ darkjobworkder --start --pid=2 // To start job worker 2
$ darkjobworkder --stop --pid=1 // To stop job worker 1
Options:
-s, --start start the buildqueue
-e, --stop stop the buildqueue
-p PID, --pid=PID specify the pid

So, I removed the urls and the values from the libimporter and made it configurable via config files. The config files are currently present in the configs/ directory. This will eventually help to allow us to add different job queues for different secondary architectures.

According to the Darkserver wiki, the steps for starting the darkproducer is:

$ darkproducer start KOJI_BUILD_ID

Now, after removing the hardlinks and making it accesible through config files, the steps to start the darkproducer is:

$ darkproducer start KOJI_BUILD_ID --config=/path/to/config/file

The configuration file for koji should be in form of:

[darkserver]
url = 'http://koji.fedoraproject.org/kojihub/'

In case of violation, the config option defaults to /etc/darkserver/darkserverurl-koji.conf

Next, I would be aiming for implementing the support for the secondary architectures, arm, ppc and add support to maintain different job queues for the different secondary architectures.

With the beginning of the 2nd week of June, I started to hack around with the project, Darkserver. The project is available on GitHub.

That week was pretty much a reading week rather than coding week as I had to go through all kinds of documentation. Darkserver is implemented using retask, which uses Redis. Though, I earlier little knowledge about the topics, I went thoroughly through the Redis documentation and googled more about Redis.

I also had to go through the retask documentation. Finally, I went once again went through darkserver code and created diagrams, this will me helping out the solve out the task and helping me to know where to hack the code to get the work running.

I will be working on Darkserver Improvement under Fedora. One can see a draft of my proposal here. My mentor is Kushal Das and i am really helpful for helping me out in the whole process both in code and application.

I was never a fan of this quote, until, I realized “The Joker” parted the right message, not for all cases but Yes this quote does hold true for certain instances.

Choose your students wisely. :P

The summer training is conducted each year during the month of June. Though, we cannot call June as the summer season but end of June is a suitable time for most of the Indian students. During this summer training, a number of upstream developers give talk on various open-source technologies. This guides the student on how to contribute back to open source. At the end of the summer training the student get a project which gives a strong base of the language as well as programming skills.

Speaking on a personal note, I did this training twice, first in 2010, then in 2011 and this training helped me pave my career path. It gave me a lot of exposure. It injected in me the habit of self-dependence, googling out your own problems, hungry for knowledge of new ideas etc. Being a student of a private engineering college, where project for any student meant any kind of online management system, this training helped me think beyond that and go for solving real life problems.

The details of the training can be found at dgplug summer training or you can shoot a mail at kushaldas AT gmail DOT com for more details.