Step 2: Grant IAM Access via Tags<\/h4>\n\n
Add an IAM policy binding that grants users access to resources with your specific tag<\/a>. The Define a ClusterRole<\/a> that specifies the exact permissions needed:<\/p>\n\n Note: While you could use a namespace-scoped Create a Apply both configurations using This approach creates a two-layer security model:<\/p>\n The result is a secure, maintainable solution that grants teams the access they need without compromising the security of other services in your cluster.<\/p>\n","pubDate":"Sun, 28 Dec 2025 00:00:00 -0600","link":"\/tech\/kubernetes-namespace-exec-access-for-external-users.html","guid":"\/tech\/kubernetes-namespace-exec-access-for-external-users.html","category":"Tech"},{"title":"FOSS contributions in 2019","description":" Heyo,<\/p>\n\n I have been interested in the concept of Freedom - both in the technical and\nsocial ecosystems for almost a decade now. Even though I am not a harcore\ncontributor or anything, I have been involved in it for few years now - as an\nenthusiast, a contributor, a mentor, and above all an evangelist. Since 2019 is\ncoming to an end, I thought I will note down what all I did last year as a FOSS\nperson.<\/p>\n\n My job at GitLab is that of a Distribution Engineer. In simple terms, I have to\ndeal with anything that a user\/customer may use to install or deploy GitLab. My\nteam maintains the omnibus-gitlab packages for various OSs, docker image, AWS\nAMIs and Marketplace listings, Cloud Native docker images, Helm charts for\nKubernetes, etc.<\/p>\n\n My job description is essentially dealing with the above mentioned tasks\nonly, and as part of my day job I don\u2019t usually have to write and backend\nRails\/Go code. However, I also find GitLab as a good open source project and\nhave been contributing few features to it over the year. Few main reasons I\nstarted doing this are<\/p>\n So, here are the features\/enhancements I implemented in GitLab, as an interested\ncontributor in the selfish interest of improving my Rails understanding and to\nget features that I wanted without much waiting:<\/p>\n\n I have been a volunteer at Swathanthra Malayalam Computing for almost 8 years\nnow. Most of my contributions are towards various localization efforts that SMC\ncoordinates. Last year, my major contributions were improving our fonts build\nprocess to help various packaging efforts (well, selfish reason - I wanted my\nlife as the maintainer of Debian packages to be easier), implementing CI based\nworkflows for various projects and helping in evangelism.<\/p>\n\n I have been a Debian contributor for almost 8 years, became a Debian Maintainer\n3 years after my first stint with Debian, and have been a Debian Developer for 2\nyears. My activities as a Debian contributor this year are:<\/p>\n\n In addition to the main projects I am a part of, I contributed to few FOSS last\nyear, either due to personal interest, or as part of my job. They are:<\/p>\n\n I also gave few talks regarding various FOSS topics that I am\ninterested\/knowledgeable in during 2019. List and details can be found at the\ntalks<\/a> page.<\/p>\n\n Overall, I think 2019 was a good year for the FOSS person in me. Next year, I\nplan to be more active in Debian because from the above list I think that is\nwhere I didn\u2019t contribute as much as I wanted.<\/p>\n","pubDate":"Wed, 01 Jan 2020 00:00:00 -0600","link":"\/tech\/foss-contributions-in-2019.html","guid":"\/tech\/foss-contributions-in-2019.html","category":"FOSS"},{"title":"DebUtsav Kochi 2018","description":" Heyo,<\/p>\n\n Been quite some time since I wrote about anything. This time, it is Debutsav.\nWhen it comes to full-fledged FOSS conferences, I usually am an attendee or at\nmost a speaker. I have given some sporadic advices and suggestions to few in the\npast, but that was it. However, this time I played the role of an organizer.<\/p>\n\n DebUtsav Kochi is the second edition of Debian Utsavam, the celebration of Free\nSoftware by Debian community. We didn\u2019t name it MiniDebConf because it was our\nrequirement for the conference to be not just Debian specific, but should\ninclude general FOSS topics too. This is specifically because our target\naudience aren\u2019t yet Debian-aware to have a Debian-only event. So, DebUtsav Kochi\nhad three tracks - one for general FOSS topics, one for Debian talks and one for\nhands-on workshops.<\/p>\n\n As a disclaimer, the description about the talks below are what I gained from my\ninteraction with the speakers and attendees, since I wasn\u2019t able to attend as\nmany talks as I would\u2019ve liked, since I was busy with the organizing stuff.<\/p>\n\n The event was organized by Free Software Community of India, whom I represented\nalong with Democratic Alliance for Knowledge Freedom (DAKF) and Student\nDeveloper Society (SDS). Cochin University of Science and Technology were\ngenerous enough to be our venue partners, providing us with necessary\ninfrastructure for conducting the event as well as accommodation for our\nspeakers.<\/p>\n\n The event span across two days, with a registration count around 150\nparticipants. Day 1 started with a keynote session by Aruna Sankaranarayanan<\/a>,\naffiliated with OpenStreetMap. She has been also associated with GNOME Project,\nWikipedia and Wikimedia Commons as well as was a lead developer of the Chennai\nFlood Map that was widely used during the floods that struck city of Chennai.<\/p>\n\n Sruthi Chandran, Debian Maintainer from Kerala, gave a brief introduction about\nthe Debian project, its ideologies and philosophies, people behind it, process\ninvolved in the development of the operating system etc. An intro about\nDebUtsav, how it came to be, the planning and organizations process that was\ninvolved in conducting the event etc were given by SDS members.<\/p>\n\n After these common talks, the event was split to two parallel tracks - FOSS and\nDebian.<\/p>\n\n In the FOSS track, the first talk was by Prasanth Sugathan of Software Freedom\nLaw Centre about the needs of Free Software licenses and ensuring license\ncompliance by projects. Parallely, Raju Devidas discussed about the process\nbehind becoming an official Debian Developer, what does it mean and why it\nmatters to have more and more developers from India etc.<\/p>\n\n After lunch, Ramaseshan S introduced the audience to Project Vidyalaya, a free\nsoftware solution for educational institutions to manage and maintain their\ncomputer labs using FOSS solutions rather than the conventional proprietary\nsolutions. Shirish Agarwal shared general idea about various teams in Debian and\nhow everyone can contribute to these teams based on their interest and ability.<\/p>\n\n Subin S showed introduced some nifty little tools and tricks that make Linux\ndesktop cool, and improve the productivity of users. Vipin George shared about\nthe possibility of using Debian as a forensic workstation, and how it can be\nmade more efficient than the proprietary counterparts.<\/p>\n\n Ompragash V from RedHat talked about using Ansible for automation tasks, its\nadvantages over similar other tools etc. Day 1 ended with Simran Dhamija talking\nabout Apache SQOOP and how it can be used for data transformation and other\nrelated usecases.<\/p>\n\n In the afternoon session of Day 1, two workshops were also conducted parallel\nto the talks. First one was by Amoghavarsha about reverse engineering, followed\nby an introduction to machine learning using Python by Ditty.<\/p>\n\n We also had an informal discussion with few of the speakers and participants\nabout Free Software Community of India, the services it provide and how to get\nmore people aware of such services and how to get more maintainers for them etc.\nWe also discussed the necessity of self-hosted services, onboarding users\nsmoothly to them and evangelizing these services as alternatives to their\nproprietary and privacy abusing counterparts etc.<\/p>\n\n Day 2 started with a keynote session by Todd Weaver, founder and CEO of Purism\nwho aims at developing laptops and phones that are privacy focused. Purism also\ndevelops PureOS, a Debian Derivative that consists of Free Software only, with\nfurther privacy enhancing modifications.<\/p>\n\n On day 2, the Debian track focused on a hands-on packaging workshop by Pirate\nPraveen and Sruthi Chandran that covered the basic workflow of packaging, the\nflow of packages through various suites like Unstable, Testing and Stable,\nstructure of packages. Then it moved to the actual process of packaging by\nguiding the participants through packaging a javascript module that is used by\nGitLab package in Debian. Participants were introduced to the tools like\nnpm2deb, lintian, sbuild\/pbuilder etc. and the various debian specific files and\ntheir functionalities.<\/p>\n\n In the FOSS track, Biswas T shared his experience in developing keralarescue.in,\na website that was heavily used during the Kerala Floods for effective\ncollaboration between authorities, volunteers and public. It was followed by\nAmoghavarsha\u2019s talk on his journey from Dinkoism to Debian. Abhijit AM of COEP\ntalked about how Free Software may be losing against Open Source and why that\nmay be a problem. Ashish Kurian Thomas shed some knowledge on few *nix tools\nand tricks that can be a productivity booster for GNU\/Linux users. Raju and\nShivani introduced Hamara Linux to the audience, along with the development\nprocess and the focus of the project.<\/p>\n\n The event ended with a panel discussion on how Debian India should move forward\nto organize itself properly to conduct more events, spread awareness about\nDebian and other FOSS projects out there, prepare for a potential DebConf in\nIndia in the near future etc.<\/p>\n\n The number of registrations and enthusiasms of the attendees for the event is\ngiving positive signs on the probability of having a proper MiniDebConf in\nKerala, followed by a possible DebConf in India, for which we have bid for.\nThanks to all the participants and speakers for making the event a success.<\/p>\n\n Thanks to FOSSEE, Hamara Linux and GitLab for being sponsors of the event and\nthus enabling us to actually do this. And also to all my co-organizers.<\/p>\n\n A very special thanks to Kiran S Kunjumon, who literally did 99% of the work\nneeded for the event to happen (as you may recall, I am good at sitting on a\nchair and planning, not actually doing anything. :D ).<\/p>\n\n Heyo,<\/p>\n\n So I attended my first international FOSS conference - FOSSAsia 2018<\/a>\nat Lifelong learning institute, Singapore<\/a>. I presented a\ntalk titled \u201cOmnibus - Serve your dish on all the tables\u201d\n(slides<\/a>, video<\/a>)\nabout the tool Chef Omnibus which I use on a daily basis for my job at\nGitLab<\/a>.<\/p>\n\n The conference was a 4-day long one and my main aim was to network with as many\npeople as I can. Well, I planned to attend sessions, but unlike earlier times\nwhen I attended all the sessions, these days I am more focussed on certain\ntopics and technologies and tend to attend sessions on those (for example,\ndevops is an area I focuses on, block chain isn\u2019t).<\/p>\n\n One additional task I had was attend the Debian booth at the exhibition from\ntime to time. It was mainly handled by Abhijith (who is a DM). I also met two\nother Debian Developers there - Andrew Lee<\/a>(alee)\nand H\u00e9ctor Or\u00f3n Mart\u00ednez<\/a>(zumbi).<\/p>\n\n I also met some other wonderful people at FOSSAsia, like Chris Aniszczyk of\nCNCF, Dr Graham Williams of Microsoft, Frank Karlitschek of NextCloud,\nJean-Baptiste Kempf and Remi Denis-Courmont of VideoLan, Stephanie Taylor of\nGoogle, Philip Paeps(trouble) of FreeBSD, Harish Pillai of RedHat, Anthony,\nChristopher Travers, Vasudha Mathur of KDE, Adarsh S of CloudCV (and who is from\nMEC College, which is quite familiar to me), Tarun Kumar of Melix, Roy Peter of\nGo-Jek (with whom I am familiar, thanks to the Ruby conferences I attended),\nDias Lonappan of Serv and many more. I also met with some whom I know knew only\ndigitally, like Sana Khan who was (yet another, :D) a Debian contributor from\nCOEP. I also met with some friends like Hari, Cherry, Harish and Jackson.<\/p>\n\n My talk went ok without too much of stuttering and I am kinda satisfied by it.\nThe only thing I forgot is to mention during the talk that I had stickers (well,\nI later placed them in the sticker table and it disappeared within minutes. So\nthat was ok. ;))<\/p>\n\n PS: Well, I had to cut down quite a lot of my explanation and drop my demo due\nto limited time. This caused me miss many important topics like omnibus-ctl or\ncookbooks that we use at GitLab. But, I had a few participants come up and meet\nme after the talk, with doubts regarding omnibus and its similarity with\nflatpak, relevance during the times of Docker etc, which was good.<\/p>\n\n Some photos are here:<\/p>\n\n Heyo,<\/p>\n\n It has been long since I have written somewhere. In the last year I attended some events, like FOSSMeet, DeccanRubyConf, GitLab\u2019s summit and didn\u2019t write anything about it. The truth is, I forgot I used to write about all these and never got the motivation to do that.<\/p>\n\n Anyway, last week, I conducted a workshop on Git basics for the students of CUSAT<\/a>. My real plan, as always, was to do a bit of FOSS evangelism too. Since the timespan of workshop was limited (10:00 to 13:00), I decided to keep everything to bare basics.<\/p>\n\n Started with an introduction to what a VCS is and how it became necessary. As a prerequisite, I talked about FOSS, concept of collaborative development, open source development model etc. It wasn\u2019t easy as my audience were not only CS\/IT students, but those from other departments like Photonics, Physics etc. I am not sure if I was able to help them understand the premise clearly. However, then I went on to talk about what Git does and how it helps developers across the world.<\/p>\n\n IIRC, this was the first talk\/workshop I did without a slide show. I was damn lazy and busy to create one. I just had one page saying \u201cGit Workshop\u201d and my contact details. So guess what? I used a whiteboard! I went over the basic concepts like repositories, commits, staging area etc and started with the hand-on session. In short, I talked about the following<\/p>\n I wanted to talk about clone, fork, branch and MRs, but time didn\u2019t permit. We wound up the session with Athul and Kiran talking about how they need the students to join the FOSSClub of CUSAT, help organizing similar workshops and how it can help them as well. I too did a bit of \u201cmotivational talk\u201d regarding how community activities can help them get a job, based on my personal experience.<\/p>\n\n Here are a few photos, courtesy of Athul and Kiran:<\/p>\n\n Heyo,<\/p>\n\n So, M.Tech is coming to an end I should probably start searching for a job soon. Still, it seems I will be having a bit of free time from Mid-September. I have got some plans about the areas I should contribute to SMC\/Indic Project. As of now, the bucket list is as follows:\n<\/p>\n\n Knowing me pretty much better than anyone else, I understand there is every chance of this being \u201cNever-being-implemented-plan\u201d (\u0d05\u0d24\u0d3e\u0d2f\u0d24\u0d4d \u0d06\u0d30\u0d02\u0d2d\u0d36\u0d42\u0d30\u0d24\u0d4d\u0d35\u0d02 :D) but still I intend to do this in an easy-first order.<\/p>\n","pubDate":"Tue, 06 Sep 2016 23:47:17 -0500","link":"\/tech\/smc_todolist.html","guid":"\/tech\/smc_todolist.html","category":["foss","smc","indicproject","community","SMC"]},{"title":"GSoC Final Report","description":" Heyo,<\/p>\n\n It is finally the time to wind up the GSoC work on which I have been buried for the past three months. First of all, let me thank Santhosh, Hrishi and Vasudev for their help and support. I seem to have implemented, or at least proved the concepts that I mentioned in my initial proposal<\/a>. A spell checker that can handle inflections in root word and generate suggestion in the same inflected form and differentiate between spelling mistakes and intended modifications has been implemented. The major contributions that I made were to<\/p>\n\n My initial work was on improving the existing stemmer that was available as part of LibIndic. The existing implementation was a rule based one that was capable of handling single levels of inflections. The main problems of this stemmer were<\/p>\n\n The above mentioned issues were fixed. The remaining category is verbs which need more detailed analysis.<\/p>\n<\/div>\n\n I too decided to maintain the rule-based approach for lemmatizer (actually, what we are designing is half way between a stemmer and lemmatizer. Since it is more inclined towards a lemmatizer, I am going to call it that.) mainly because for implementing any ML or AI techniques, there should be sufficient training data, without which the efficiency will be very poor. It felt better to gain higher efficiency with available rules than to try out ML techniques with no guarantee (Known devil is better logic).<\/p>\n\n The basic logic behind the multi-level inflection handling lemmatizer is iterative suffix stripping. At each iteration, a suffix is identified from the word and it is transformed to something else based on predefined rules. When no more suffixes are found that have a match on the rule set, we assume the multiple levels of inflection have been handled.<\/p>\n\n To handle root words that look like inflected words (hereafter called \u2018exceptional words\u2019) from being stemmed unnecessarily, it is obvious we have to use a root word corpus. I used the Datuk dataset that is made available openly by Kailash as the root word corpus. A corpus comparison was performed before the iterative suffix stripping started, so as to handle root words without any inflection. Thus, the word \u0d06\u0d2a\u0d24\u0d4d\u0d24\u0d4d will get handled even before the iteration begins. However, what if the input word is an inflected form of an exceptional word, like \u0d06\u0d2a\u0d24\u0d4d\u0d24\u0d3f\u0d32\u0d47\u0d15\u0d4d\u0d15\u0d4d? This makes it necessary to introduce the corpus comparison step after each iteration.<\/p>\n\n At each iteration, suffix stripping happens from left to right. Initial suffix has 2nd character as the starting point and last character as end point. At each inner iteration, the starting point moves rightwards, thus making the suffix shorter and shorter. Whenever a suffix is obtained that has a transformation rule defined in the rule set, it is replaced with the corresponding transformation. This continues until the suffix becomes null.<\/p>\n\n Multi-level inflection is handled on the logic that each match in rule set induces a hope that there is one more inflection present. So, before each iteration, a flag is set to False. Whenever a match in ruleset occurs at that iteration, it is set to true. If at the end of an iteration, the flag is true, the loop repeats. Else, we assume all inflections have been handled.<\/p>\n\n Since this lemmatizer is also used along with a spellchecker, we will need a history of the inflections identified so that the lemmatization process can be reversed. For this purpose, I tagged the rules unambiguously. Each time an inflection is identified, that is the extracted suffix finds a match in the rule set, in addition to the transformation, the associated tag is also pushed to a list. As the result, the stem along with this list of tags is given to the user. This list of tags can be used to reverse the lemmatization process - for which I wrote an inflector function.<\/p>\n\n A demo screencast of the lemmatizer is given below.<\/p>\n\nKubernetes Engine Viewer<\/code> role (roles\/container.viewer<\/code>) provides sufficient base permissions without granting excessive access.<\/p>\n\nStep 3: Create a Kubernetes ClusterRole<\/h4>\n\n
apiVersion<\/span>:<\/span> rbac.authorization.k8s.io\/v1<\/span>\nkind<\/span>:<\/span> ClusterRole<\/span>\nmetadata<\/span>:<\/span>\n name<\/span>:<\/span> custom-access-role<\/span>\nrules<\/span>:<\/span>\n -<\/span> apiGroups<\/span>:<\/span> [<\/span>\"<\/span>\"<\/span>]<\/span>\n resources<\/span>:<\/span> [<\/span>\"<\/span>pods\"<\/span>,<\/span> \"<\/span>pods\/exec\"<\/span>,<\/span> \"<\/span>pods\/attach\"<\/span>,<\/span> \"<\/span>pods\/portforward\"<\/span>,<\/span> \"<\/span>pods\/log\"<\/span>]<\/span>\n verbs<\/span>:<\/span> [<\/span>\"<\/span>get\"<\/span>,<\/span> \"<\/span>list\"<\/span>,<\/span> \"<\/span>watch\"<\/span>,<\/span> \"<\/span>create\"<\/span>]<\/span>\n<\/code><\/pre><\/div><\/div>\n\nRole<\/code>, a ClusterRole<\/code> offers better reusability if you need similar permissions for other namespaces later.<\/p>\n\nStep 4: Bind the Role to Users<\/h4>\n\n
RoleBinding<\/code> to connect the role to specific users and namespaces:<\/p>\n\napiVersion<\/span>:<\/span> rbac.authorization.k8s.io\/v1<\/span>\nkind<\/span>:<\/span> RoleBinding<\/span>\nmetadata<\/span>:<\/span>\n name<\/span>:<\/span> custom-rolebinding<\/span>\n namespace<\/span>:<\/span> my-namespace<\/span>\nsubjects<\/span>:<\/span>\n -<\/span> kind<\/span>:<\/span> User<\/span>\n name<\/span>:<\/span> myuser@gmail.com<\/span>\n apiGroup<\/span>:<\/span> rbac.authorization.k8s.io<\/span>\nroleRef<\/span>:<\/span>\n kind<\/span>:<\/span> ClusterRole<\/span>\n name<\/span>:<\/span> custom-access-role<\/span>\n apiGroup<\/span>:<\/span> rbac.authorization.k8s.io<\/span>\n<\/code><\/pre><\/div><\/div>\n\nkubectl apply -f <filename><\/code>.<\/p>\n\nHow It Works<\/h3>\n\n
\n
GitLab<\/h2>\n\n
\n
Accepting Merge Requests<\/code> label on them, which meant Product was in\nagreement that the feature was worthy of having, but there were issues with\nmore priority to be tackled first.<\/li>\n<\/ol>\n\n\n
Swathanthra Malayalam Computing<\/h2>\n\n
\n
Debian<\/h2>\n\n
\n
fonts-smc-*<\/code> and hyphen-indic<\/code> packages.<\/li>\n Other FOSS projects<\/h2>\n\n
\n
\n
Talks<\/h2>\n\n
![\"Group](\"\/images\/tech\/debutsav\/photo.jpg\")
<\/p>\n","pubDate":"Wed, 19 Dec 2018 00:00:00 -0600","link":"\/tech\/debutsav.html","guid":"\/tech\/debutsav.html","category":"debian"},{"title":"FOSSAsia 2018 - Singapore","description":"![\"Abhijith](\"\/images\/tech\/fossasia\/Abhijith.jpg\")
<\/p>\n![\"Abhijith](\"\/images\/tech\/fossasia\/Abhijith_VLC.jpg\")
<\/p>\n![\"Andrew's](\"\/images\/tech\/fossasia\/alee.jpg\")
<\/p>\n![\"With](\"\/images\/tech\/fossasia\/Anthony_Harish.jpg\")
<\/p>\n![\"Chris](\"\/images\/tech\/fossasia\/Chris.jpg\")
<\/p>\n![\"Debian](\"\/images\/tech\/fossasia\/debian-booth.jpg\")
<\/p>\n![\"Frank](\"\/images\/tech\/fossasia\/Frank.jpg\")
<\/p>\n![\"Graham](\"\/images\/tech\/fossasia\/Graham.jpg\")
<\/p>\n![\"MOS](\"\/images\/tech\/fossasia\/MOS.jpg\")
<\/p>\n![\"Premas](\"\/images\/tech\/fossasia\/Premas.jpg\")
<\/p>\n![\"The](\"\/images\/tech\/fossasia\/Premas_ML.jpg\")
<\/p>\n![\"With](\"\/images\/tech\/fossasia\/sana.jpg\")
<\/p>\n![\"Well,](\"\/images\/tech\/fossasia\/Tamil.jpg\")
<\/p>\n![\"Zumbi's](\"\/images\/tech\/fossasia\/zumbi.jpg\")
<\/p>\n\n
![\"\"](\"\/images\/tech\/cusat-git\/image_1.jpg\")
<\/p>\n\n![\"\"](\"\/images\/tech\/cusat-git\/image_2.jpg\")
<\/p>\n\n![\"\"](\"\/images\/tech\/cusat-git\/image_3.jpg\")
<\/p>\n\n![\"\"](\"\/images\/tech\/cusat-git\/image_4.jpg\")
<\/p>\n","pubDate":"Wed, 17 Jan 2018 00:00:00 -0600","link":"\/tech\/git-intro-workshop-cusat.html","guid":"\/tech\/git-intro-workshop-cusat.html"},{"title":"SMC\/IndicProject Activities- ToDo List","description":"\n
\n
1. Lemmatizer\/Stemmer<\/h2>\n\n
TLDR<\/h3>\n\n
\n
![\"Lemmatizer](\"\/images\/tech\/lemmatizer_flowchart.png\")
<\/p>\n\n