<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Gurpreet Singh Kambo on Medium]]></title>
        <description><![CDATA[Stories by Gurpreet Singh Kambo on Medium]]></description>
        <link>https://medium.com/@gurpreet.devops?source=rss-3f9fab711112------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*CYlcE9FGdj1avz6P6unD5A.jpeg</url>
            <title>Stories by Gurpreet Singh Kambo on Medium</title>
            <link>https://medium.com/@gurpreet.devops?source=rss-3f9fab711112------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 09:22:49 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@gurpreet.devops/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Docker Essential Commands]]></title>
            <link>https://medium.com/@gurpreet.devops/docker-essential-commands-4b382a529f42?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/4b382a529f42</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[containers]]></category>
            <category><![CDATA[docker-command]]></category>
            <category><![CDATA[docker]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Mon, 09 Nov 2026 16:42:40 GMT</pubDate>
            <atom:updated>2026-11-09T16:42:40.586Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RAQgIpvdMc9cq6l32CA74A.gif" /></figure><p>“It works on my machine.” “Great, we’ll ship your machine.” And then we actually did.</p><p>I remember when “just put it in a container” started replacing whole paragraphs of deployment documentation. It felt like magic for about eighteen months. Then I learned containers don’t remove the hard problems, they relocate them.</p><p>Still, Docker earned its spot. These are the commands I use most, in the order I use them:</p><p>🏗️ 𝐁𝐔𝐈𝐋𝐃<br>▸ docker build -t myapp:1.0 . — build an image<br>▸ docker build — no-cache -t myapp . — ignore cache<br>▸ docker images — list images<br>▸ docker rmi myapp:1.0 — remove image</p><p>▶️ 𝐑𝐔𝐍<br>▸ docker run -d -p 8080:80 — name web nginx — background + map a port<br>▸ docker run — rm -it ubuntu bash — throwaway shell<br>▸ docker run -e ENV=prod -v $(pwd):/app myapp — env var + mount a folder<br>▸ docker ps -a — all containers</p><p>📜 𝐋𝐎𝐆𝐒 &amp; 𝐈𝐍𝐒𝐏𝐄𝐂𝐓<br>▸ docker logs -f web — follow logs<br>▸ docker logs — tail 100 web — last 100 lines<br>▸ docker inspect web — full JSON<br>▸ docker stats — live CPU / memory</p><p>🚪 𝐄𝐗𝐄𝐂 &amp; 𝐂𝐎𝐏𝐘<br>▸ docker exec -it web sh — open a shell<br>▸ docker exec web ls /usr/share/nginx/html — run one command inside<br>▸ docker cp web:/etc/nginx/nginx.conf . — copy a file out<br>▸ docker top web — processes inside</p><p>📦 𝐓𝐀𝐆 &amp; 𝐏𝐔𝐒𝐇<br>▸ docker login — authenticate<br>▸ docker tag myapp:1.0 user/myapp:1.0 — name it for the registry<br>▸ docker push user/myapp:1.0 — upload image<br>▸ docker pull user/myapp:1.0 — download image</p><p>🧹 𝐂𝐎𝐌𝐏𝐎𝐒𝐄 &amp; 𝐂𝐋𝐄𝐀𝐍𝐔𝐏<br>▸ docker compose up -d — start the stack<br>▸ docker compose down — stop + remove<br>▸ docker system prune -f — remove unused data<br>▸ docker volume ls — list volumes</p><p>Containers are a great tool. They are not a substitute for understanding your own system.</p><p>Wrapping a messy garage in a nicer box doesn’t count as cleaning it.</p><p>What’s a tool that solved one real problem for you and got credit for five it never touched?</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4b382a529f42" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Canadian Job Market is Tough!]]></title>
            <link>https://medium.com/@gurpreet.devops/canadian-job-market-is-tough-cac4d3c03b65?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/cac4d3c03b65</guid>
            <category><![CDATA[canada]]></category>
            <category><![CDATA[canadian-job-market]]></category>
            <category><![CDATA[canada-jobs]]></category>
            <category><![CDATA[it-job-market]]></category>
            <category><![CDATA[jobs]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Wed, 04 Nov 2026 16:15:59 GMT</pubDate>
            <atom:updated>2026-11-04T16:15:59.555Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/928/1*qDk2tEeU4E-05zzWPux3vQ.png" /></figure><p>The Canadian job market is tough right now, and I’m not going to pretend otherwise.</p><p>Every week, another layoff announcement. Another friend updating their profile to “Open to Work.” Another group chat that starts with “Did you hear that?”</p><p>I’ve been in tech for 23 years, and I moved to Canada as an international student with no network and zero idea how a Toronto winter worked.</p><p>I’ve also been on both sides of a layoff process. So I’ll say what I wish someone had told me: a hard market feels personal, and it isn’t. It’s a market.</p><p>You can’t always prevent layoffs. You can choose how people are treated on the way out, and Canva just showed what that looks like.</p><p>Here’s what they’re offering everyone affected:<br>- At least six months of salary, in every location<br>- Twelve months of healthcare and family-inclusive wellbeing support<br>- Career coaching, interview prep and a year of LinkedIn Premium<br>- Keeping their MacBooks, because not everyone has a personal laptop<br>- A US$10,000 fund for anyone starting something of their own</p><p>That laptop line stuck with me. Someone who just lost their job shouldn’t also have to wonder how they’ll update their resume or show up to their next interview. Kudos to <a href="https://www.linkedin.com/company/canva/">Canva</a> for an example worth learning from.</p><p>It’s also a useful checklist if you’re ever on the receiving end. Here’s what I’d tell a friend:<br>- Look past severance. Ask about healthcare continuation, career support, equipment and any transition fund. Canva’s list shows these things exist.</p><p>- Make your resume do the work in six seconds. Put your biggest result in the first lines, with a real number if you have one.</p><p>- Referrals beat applications. A short message to someone who knows you gets read. A 200th “Easy Apply” mostly gets filed under “later.”</p><p>-Talk to people before you need to. Coffee chats in a good month cost nothing. In a bad month they’re gold.</p><p>- Keep consulting and contract work on the table. It pays bills, keeps your skills sharp and often turns into the permanent role.</p><p>- Learn what the job posts keep asking for. Right now that’s cloud, platform engineering, FinOps and using AI in real work.</p><p>- Protect your energy. Apply in the morning, then go for a walk. Rejection piles up faster when you’re not sleeping or eating properly.</p><p>And if you’re not job hunting today, you may be standing next to someone who is. Send the referral. Forward the posting. Reply to the “I got laid off” post with something kinder than “thoughts and prayers.”</p><p>This market will turn. It always does, usually right after we’ve all stopped checking. Until then, be kind to yourself.</p><p>Keep up the hustle!</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cac4d3c03b65" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Git Essential Commands]]></title>
            <link>https://medium.com/@gurpreet.devops/git-essential-commands-3a9c9b7b34fb?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/3a9c9b7b34fb</guid>
            <category><![CDATA[github]]></category>
            <category><![CDATA[basic-git-commands]]></category>
            <category><![CDATA[git-commands]]></category>
            <category><![CDATA[git]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Mon, 02 Nov 2026 17:27:13 GMT</pubDate>
            <atom:updated>2026-11-02T17:27:13.447Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9OxRGIiQQ4ipsM3KmeEpyg.gif" /></figure><p>My commit history says “fix”, “fix2”, “fix final” and “fix final REAL”. Yours does too!</p><p>23 years in, Git is still the one tool every engineer uses daily and nobody fully trusts. We all know someone (okay, it’s us) who has typed git reset — hard and then stared at the screen like it owed an apology.</p><p>Here are the commands I actually use, in the order a change travels from idea to production:</p><p>📥 𝐂𝐋𝐎𝐍𝐄 &amp; 𝐂𝐎𝐍𝐅𝐈𝐆<br>▸ git init — new repo<br>▸ git clone <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd%2Ein%2FgFMSc6nq&amp;urlhash=kMHd&amp;mt=gwti8JhJmqhKs80C92tGbm5QUbHIF4NjQ4VvEn9TDroVPamybOHWLnOHCNgZZuqrRMljtQxtGd6q-baGeEH6LiSyyYiybhn4RQ-1in4cnNFTecEXjArior3DoJSWMnSKKW6npy4drAq3IFy4m0Da-IdU-giytAJSR8xvzB-8sZssJyIOTkAtcR64FJAf7AeN&amp;isSdui=true">https://lnkd.in/gFMSc6nq</a> — copy a remote repo<br>▸ git clone — depth 1 &lt;url&gt; — shallow clone<br>▸ git config — global <a href="https://www.linkedin.com/safety/go/?url=http%3A%2F%2Fuser%2Ename&amp;urlhash=u6e4&amp;mt=OlCjHoB-B-PEwcL9oEvybshwQraC7D9KE1BdZ2Mciwag6ArfCpyf9yiFTrAaWA_3kQsEEMPLWgmrqfT8A2__RUIreHA2rKRkVCoAcssAV0qQ6bMgP9MYs1KNxZ0t8rWW8rrNZUiBzUmboo2OJGXhyggfvchUXn8PquwuY-vnxnG1MvQAL1ZdGTJVdNKo5HKQ&amp;isSdui=true">user.name</a> “Your Name” — set your identity</p><p>🌿 𝐁𝐑𝐀𝐍𝐂𝐇<br>▸ git switch -c feature/login — create + switch<br>▸ git branch -a — all branches<br>▸ git switch main — go back<br>▸ git branch -d feature/login — delete merged</p><p>💾 𝐒𝐓𝐀𝐆𝐄 &amp; 𝐂𝐎𝐌𝐌𝐈𝐓<br>▸ git status -s — short status<br>▸ git add -p — stage hunks<br>▸ git commit -m “Add login form” — save with a message<br>▸ git commit — amend — no-edit — fix last commit</p><p>🔄 𝐏𝐔𝐒𝐇 &amp; 𝐏𝐔𝐋𝐋<br>▸ git push -u origin feature/login — publish + track branch<br>▸ git pull — rebase origin main — update, keep history tidy<br>▸ git fetch — all — prune — refresh remotes<br>▸ git remote -v — show remotes</p><p>🔀 𝐌𝐄𝐑𝐆𝐄 &amp; 𝐑𝐄𝐁𝐀𝐒𝐄<br>▸ git merge feature/login — merge into current<br>▸ git rebase main — replay on main<br>▸ git cherry-pick 1a2b3c4 — copy one commit<br>▸ git stash push -m “wip” — shelve changes</p><p>🔍 𝐋𝐎𝐆 · 𝐃𝐈𝐅𝐅 · 𝐔𝐍𝐃𝐎<br>▸ git log — oneline — graph -10 — history<br>▸ git diff — staged — to be committed<br>▸ git restore — staged <a href="https://www.linkedin.com/safety/go/?url=http%3A%2F%2Fapp%2Epy&amp;urlhash=ioea&amp;mt=O5p1j0vlzLncvTURV0FpeSQazapikv16eXadebOvDoepZH5LiXdVSeqyfAAIqspcN75Q8rEd7gtsgTGvwwk9kco7uqdVbCiKmM5wzH1SL8413erSYVCkTct9JrAdgco2-lMka51T0fHKqTagtyPgdmIlOtNrRz3wk7pBVIdgzzRpd2ZyDkSW2QNOcaaZCiqS&amp;isSdui=true">app.py</a> — unstage<br>▸ git reset — hard HEAD~1 — discards work!</p><p>My rule: run git status before anything with the word “hard” or “force” in it. A commit message is a note to the 2 AM version of someone who has never seen your code, so write it for them.</p><p>What’s the best (worst) commit message you’ve ever written? Confession is welcome :)</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3a9c9b7b34fb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I’ve Been on Both Sides of a Layoff. Here’s What I’d Tell a Friend at Microsoft!]]></title>
            <link>https://medium.com/@gurpreet.devops/ive-been-on-both-sides-of-a-layoff-here-s-what-i-d-tell-a-friend-1f4d422c2153?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f4d422c2153</guid>
            <category><![CDATA[careers]]></category>
            <category><![CDATA[canada]]></category>
            <category><![CDATA[microsoft]]></category>
            <category><![CDATA[layoffs]]></category>
            <category><![CDATA[career-advice]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Thu, 29 Oct 2026 15:29:20 GMT</pubDate>
            <atom:updated>2026-10-29T15:32:02.083Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5eFd0WsLjlwuDBhmZmESnw.png" /></figure><p>Microsoft offers voluntary buyouts to 60,000 employees in the U.S., Canada, India and the United Kingdom.</p><p>Holding a buyout offer from Microsoft today? Breathe. Please hear me: you do NOT have to decide this week.</p><p>I’ve spent 23 years in tech, through reorgs, restructures and a few “exciting changes”, and nobody I’ve seen decides well in a panic. It’s voluntary, but anyone who’s been through a company shake-up knows “voluntary” can still feel like a heavy question with a short deadline.</p><p>I’ve been on both sides of a layoff process, so I know two things. The person getting the offer is usually more scared than they let on. And the people making the decisions are rarely as certain as they sound in the memo.</p><p>I’ve watched talented people go through this, and the hardest part was rarely the money. It was the sudden “who am I without this title?” feeling.</p><p>Whether you stay, take the offer, or are somewhere nearby watching it happen to people you care about, here’s what I’d tell a friend:</p><p>- Read the whole package before you feel anything about it. Severance, benefits, equity vesting, pension or retirement contributions, health coverage, and the exact last day. Then ask HR every question twice.</p><p>- Talk to someone neutral. A fee-only financial advisor or an employment lawyer is worth a few hundred dollars here. I’m not either of those, so please don’t treat this as advice.</p><p>- Check the local details. In Canada, for example, find out how a voluntary exit affects benefits and Employment Insurance before you sign, not after.</p><p>- Do the “what if” math. How many months can you comfortably cover? That number tells you whether you’re choosing or being pushed.</p><p>- Warm up your network now, while you don’t need anything from it. A coffee chat in November feels very different from an ask in January.</p><p>- Update your LinkedIn and resume this week, even if you stay. Think of it as insurance. Nobody has ever regretted having a current resume.</p><p>- Keep learning the things the market keeps rewarding: cloud, platform engineering, FinOps, and using AI well, not just talking about it.</p><p>And one more thing. Your title is something you rent, and you give it back eventually. You’re still the same good person you were before it, to the people who actually matter.</p><p>If you’re affected by this, my DMs are open. I’m happy to share what I know, review a resume, or just listen. No agenda.</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f4d422c2153" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Terraform Essential Commands]]></title>
            <link>https://medium.com/@gurpreet.devops/terraform-essential-commands-09f4d5346768?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/09f4d5346768</guid>
            <category><![CDATA[infrastructure-as-code]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[azure]]></category>
            <category><![CDATA[gcp]]></category>
            <category><![CDATA[terraform]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Mon, 26 Oct 2026 14:14:50 GMT</pubDate>
            <atom:updated>2026-10-26T14:14:50.730Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U6XnNrmHif6de3RxHmNLpg.gif" /></figure><p>Terraform apply -auto-approve is a trust fall with production.</p><p>In 2003, my “infrastructure plan” was a printed checklist and a phone call to someone named Rahul. Twenty-three years later, Terraform does the same job, except it actually shows you what’s about to change before you change it.</p><p>It’s the only meeting where someone tells you exactly what’s about to be deleted, and half of us still say “looks fine.”</p><p>I’ve used Terraform to build everything from tidy single-cloud setups to the same idea running across AWS, Azure and GCP. These are the commands I reach for, in the order I reach for them:</p><p>🚀 𝐈𝐍𝐈𝐓</p><p>▸ terraform init — download providers</p><p>▸ terraform init -upgrade — bump provider versions</p><p>▸ terraform init -reconfigure — switch backend</p><p>▸ terraform init -backend-config=backend.hcl — backend settings file</p><p>✨ 𝐅𝐌𝐓 &amp; 𝐕𝐀𝐋𝐈𝐃𝐀𝐓𝐄</p><p>▸ terraform fmt -recursive — format all .tf files</p><p>▸ terraform fmt -check — fail CI if unformatted</p><p>▸ terraform fmt -diff — show what changes</p><p>▸ terraform validate — catch syntax errors</p><p>👀 𝐏𝐋𝐀𝐍</p><p>▸ terraform plan — preview changes</p><p>▸ terraform plan -out=tfplan — save plan to a file</p><p>▸ terraform plan -var=”env=prod” — override a variable</p><p>▸ terraform plan -refresh-only — detect drift</p><p>⚡ 𝐀𝐏𝐏𝐋𝐘</p><p>▸ terraform apply tfplan — apply the saved plan</p><p>▸ terraform apply -auto-approve — skip prompt (CI/CD)</p><p>▸ terraform apply -var-file=prod.tfvars — env-specific values</p><p>▸ terraform apply -replace=aws_instance.web — force recreate</p><p>💣 𝐃𝐄𝐒𝐓𝐑𝐎𝐘</p><p>▸ terraform plan -destroy — preview what will go</p><p>▸ terraform destroy — tear down everything</p><p>▸ terraform destroy -auto-approve — no prompt, careful!</p><p>▸ terraform destroy -target=aws_instance.web — remove one resource</p><p>🗂️ 𝐒𝐓𝐀𝐓𝐄 &amp; 𝐈𝐌𝐏𝐎𝐑𝐓</p><p>▸ terraform state list — what is tracked</p><p>▸ terraform import aws_s3_bucket.logs my-logs — adopt an existing bucket</p><p>▸ terraform workspace new staging — separate env state</p><p>▸ terraform output vpc_id — read an output</p><p>My rule after 22 years of prod deployments: save the plan, read the plan, then apply the saved plan. The command that deletes something never announces itself, so find the -/+ line before it finds you.</p><p>Which Terraform command do you trust the least?</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=09f4d5346768" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[24 Linux Essential commands I still use after 22 years!]]></title>
            <link>https://medium.com/@gurpreet.devops/24-linux-essential-commands-i-still-use-after-22-years-b921840a7d68?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/b921840a7d68</guid>
            <category><![CDATA[command-line]]></category>
            <category><![CDATA[linux-commands]]></category>
            <category><![CDATA[linux]]></category>
            <category><![CDATA[devops]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Mon, 19 Oct 2026 20:39:11 GMT</pubDate>
            <atom:updated>2026-10-19T20:39:11.896Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WcqpDVQSkTqQpxok3UXRDw.gif" /></figure><p>Linux is free. The weekend you spend fixing it is not.</p><p>Linux has about 1,000 commands. I use roughly 24 of them, and that gets me through most 2 AM incidents.</p><p>I put them in a GIF so you don’t have to dig through a wiki while production is on fire.</p><p>Six groups, in the order I reach for them:</p><p>📁 𝐅𝐈𝐋𝐄𝐒<br>▸ ls -lah — long list<br>▸ cp -r src/ backup/ — copy a folder<br>▸ find . -name “*.log” -mtime +7 — find old logs<br>▸ rm -rf build/ — delete (careful!)</p><p>🔎 𝐒𝐄𝐀𝐑𝐂𝐇 &amp; 𝐓𝐄𝐗𝐓<br>▸ grep -rn “ERROR” /var/log/ — search logs<br>▸ grep -i timeout app.log | wc -l — count matches<br>▸ tail -f /var/log/syslog — follow a log<br>▸ awk ‘{print $1}’ access.log | sort | uniq -c — hits per IP</p><p>🔐 𝐏𝐄𝐑𝐌𝐈𝐒𝐒𝐈𝐎𝐍𝐒<br>▸ chmod +x <a href="https://www.linkedin.com/safety/go/?url=http%3A%2F%2Fdeploy%2Esh&amp;urlhash=3q8Z&amp;mt=4sOziwI-XKXj_cyhhfaoNvsk_6C_Rb-t7f2QjvzmIZHaYThpZ-XB-rwQ-Et-GwNKRx0h5ts7GbfSeUOUfoFC7BtxXCQAgt0fN1BKaRrPFazlgQOKDbvtBr-Y&amp;isSdui=true">deploy.sh</a> — make executable<br>▸ chmod 640 secrets.env — owner rw, group r<br>▸ chown -R www-data:www-data /var/www — change owner<br>▸ id — who am I</p><p>⚙️ 𝐏𝐑𝐎𝐂𝐄𝐒𝐒𝐄𝐒<br>▸ ps aux — sort=-%mem | head — top memory users<br>▸ kill -15 1234 — graceful stop<br>▸ systemctl status nginx — service state<br>▸ journalctl -u nginx -f — follow service logs</p><p>🌐 𝐍𝐄𝐓𝐖𝐎𝐑𝐊<br>▸ ip a — interfaces and IPs<br>▸ ss -tulpn — listening ports<br>▸ curl -I <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Fexample%2Ecom&amp;urlhash=P04_&amp;mt=PdiB7GFVSnBMnh_MvvE2s4O0LfOL96RiXEzk78rl1XD7I9MtB0jEom0k5QQC5JWz0DuZhspVlLqg2hMVdU8RITOM6wcMoagcqciQV1Z1Ux3eyFEYIAFzTLxc&amp;isSdui=true">https://example.com</a> — HTTP headers<br>▸ dig +short <a href="https://www.linkedin.com/safety/go/?url=http%3A%2F%2Fexample%2Ecom&amp;urlhash=xpOf&amp;mt=z2xjxqmP28J5zPr70OcscukJBAzd_wxrb4eVpettWTwX3EXTrXqmySXV5KymgFaHuVK0bNSiwBw9v9AQf-otuGsucX7J7iO_ATUkZAnjDH6X5006PuV45BBQ&amp;isSdui=true">example.com</a> — DNS lookup</p><p>💾 𝐃𝐈𝐒𝐊 &amp; 𝐒𝐘𝐒𝐓𝐄𝐌<br>▸ df -h — disk usage<br>▸ du -sh * | sort -h — biggest folders<br>▸ free -h — memory<br>▸ uname -a — kernel info</p><p>One habit that’s saved me more than any command: I run ls before rm. Every time. It takes two seconds and has never once been the wrong decision.</p><p>Save this for the day you SSH into a box you’ve never seen and need to look like you know what you’re doing.</p><p>Which command do you use most that never makes the cheat sheets?</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b921840a7d68" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Chef vs Puppet vs Ansible: Why I Stopped Picking a Team]]></title>
            <link>https://medium.com/@gurpreet.devops/chef-vs-puppet-vs-ansible-why-i-stopped-picking-a-team-f39460ceca46?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/f39460ceca46</guid>
            <category><![CDATA[puppet]]></category>
            <category><![CDATA[configuration-management]]></category>
            <category><![CDATA[ansible]]></category>
            <category><![CDATA[chefs]]></category>
            <category><![CDATA[devops]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Thu, 15 Oct 2026 15:46:22 GMT</pubDate>
            <atom:updated>2026-10-15T15:46:22.508Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Chef vs Puppet vs Ansible: Why I Stopped Picking a Team</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uJ0ZnANv-WBf5CCGo8M7Ow.png" /></figure><p>I’ve shipped production config changes through Chef, Puppet, and Ansible, at different employers, in different decades of my career, and I have never once seen any of the three camps fully convince the others of anything in a Slack/Teams thread.</p><p>Chef people will tell you that treating infrastructure as actual testable, versionable code — recipes, cookbooks, the whole Ruby-flavored apparatus — is the “real” engineering approach, and if you enjoy writing infrastructure the way you write application code, they have a point.</p><p>Puppet people will tell you declarative, agent-based configuration management is the mature, battle-tested way to keep fleets of servers in a known state, and they’re not wrong either.</p><p>Ansible people will tell you agentless, push-based simplicity means you can get something running in twenty minutes instead of two days of setting up an agent architecture, and they’re also not wrong.</p><p>This is the most Canadian argument in infrastructure: all three sides are being reasonable and somehow that makes it worse.</p><p>What twenty-two years of picking sides depending on the job actually taught me: the tool matched the problem, not the trend.</p><p>Chef made sense on a team that already thought like software engineers and wanted config management to feel like writing tests.</p><p>Puppet made sense for a large, stable fleet of long-lived servers with real drift-management needs.</p><p>Ansible made sense when I needed fifteen ephemeral cloud instances configured by Thursday and didn’t have time to argue about it.</p><p>I’ve used all three where each one made sense, and the one time I tried to force one tool into another’s job out of sheer personal loyalty, I regretted it by the following Monday.</p><p>Loyalty to a specific tool is a personality trait. It is not an architecture decision.</p><p>Chef, Puppet, or Ansible — and be honest about whether that’s a technical opinion or just muscle memory from your first job.</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f39460ceca46" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI is coming for my job. Again.]]></title>
            <link>https://medium.com/@gurpreet.devops/ai-is-coming-for-my-job-again-84d2afe456af?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/84d2afe456af</guid>
            <category><![CDATA[career-change]]></category>
            <category><![CDATA[motivation]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[jobs]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 15:05:09 GMT</pubDate>
            <atom:updated>2026-10-09T15:05:09.379Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/976/1*fGgLzTQP0ah-cbUpCrutNw.png" /></figure><p>AI is coming for my job. Again. I’ve lost count, but I think this is attempt number five.</p><p>When I was a kid growing up in New Delhi, people thought calculators were going to end maths teachers. My maths teacher, Joshi sir, seemed fine. They just stopped grading long division and started asking “why” questions, which were somehow worse.</p><p>Then GPS was going to make maps and directions useless. It did. And yet every family still has one person (shout-out to my Jabalpur wale Mama ji) who ignores it with full confidence: “No, no, I know a shortcut.” The shortcut has never once been shorter.</p><p>My own career has been one long “this will replace you” headline.</p><p>In 2003, at McAfee, our deployment process was a printed checklist and a phone call. Scripting arrived and was supposed to end the people doing it by hand. Guess what happened next? I became the person who wrote the scripts.</p><p>Then I moved from New Delhi to Canada, and no technology was needed to “replace” me. A new country did it for free.</p><p>Years of engineering experience, and I was sitting in a Seneca College classroom next to people much younger than me, re-explaining skills I already had.</p><p>No tool and no AI got me through that reset. People did: a professor who took a chance on me (thank you, Professor Beth), and managers who looked past where my experience came from (thank you, Max and Craig).</p><p>Even my degree is a story about this. I studied engineering, and nobody in that program ever said “Kubernetes.” The specific skills expired, but the habit of asking “what breaks first, and why?” never did. That’s the part no machine took.</p><p>So here’s my honest take on AI:</p><ul><li>It drafts the runbook nobody wanted to write. Lovely.</li><li>It once invented a Kubernetes annotation with total confidence. 100% sure, 0% right. Basically the GPS shortcut uncle.</li><li>It didn’t find the $3M I saved at BMO. A team of humans with a FinOps plan and a lot of coffee did.</li><li>It doesn’t know that in healthcare, “the system is down” means patient care is waiting. It doesn’t lose sleep over that. I do.</li></ul><p>AI will take tasks, the same way calculators took long division and GPS took the folded map in the glovebox. The job just moves up a level, toward judgment, trust and the human stuff. Every time a tool showed up to replace me, my job got more human, not less.</p><p>So I’m not worried about AI taking my job. I’m worried about people who refuse to learn it, competing with people who did.</p><p>Your turn: what’s something, at work or in life, that was supposed to replace you, and what did you become instead?</p><p>#AI #FutureOfWork #CareerGrowth #ImmigrantInTech #DevOps #Leadership</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=84d2afe456af" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Containers Didn’t Fix Our Hard Problems. They Just Gave Them a Nicer Suit.]]></title>
            <link>https://medium.com/@gurpreet.devops/containers-didnt-fix-our-hard-problems-they-just-gave-them-a-nicer-suit-cc4d7d265892?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/cc4d7d265892</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[containers]]></category>
            <category><![CDATA[docker]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Thu, 08 Oct 2026 14:42:16 GMT</pubDate>
            <atom:updated>2026-10-08T14:42:16.749Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OWG_Ut655o7wEa8H21I8Bw.png" /></figure><p>Somewhere in the early-2010s stretch of my career, “just put it in a container” started replacing entire paragraphs of deployment documentation.</p><p>I also remember the exact meeting where someone said “it works on my machine” and someone else said “great, we’ll ship your machine” and nobody laughed because we all realized that was, actually, now technically possible.</p><p>It felt like magic for about eighteen months. Then reality did what reality does: containers didn’t remove the hard problems, they just relocated them.</p><p>Instead of “why doesn’t this work on the server,” we got “why does this container behave differently in the orchestrator than it did on my laptop,” which is the same argument wearing a much more expensive suit.</p><p>Here’s what going all-in on containers across multiple employers actually taught me: the technology solved packaging and consistency beautifully.</p><p>It did not, and was never going to, solve the fact that someone still has to understand what the application actually does. I’ve watched teams containerize a mess and feel proud of themselves, as if wrapping a disorganized garage in a nicer box counts as cleaning it.</p><p>Containers are a genuinely great tool. They are not a substitute for understanding your own system. Ask me how I know.</p><p>What’s a tool that solved one real problem for you and then got credit for solving five it never touched?</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cc4d7d265892" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Quit a Decade-Long Engineering Career to Go Back to School]]></title>
            <link>https://medium.com/@gurpreet.devops/i-quit-a-decade-long-engineering-career-to-go-back-to-school-538eb1207e6f?source=rss-3f9fab711112------2</link>
            <guid isPermaLink="false">https://medium.com/p/538eb1207e6f</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[career-advice]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[writing]]></category>
            <dc:creator><![CDATA[Gurpreet Singh Kambo]]></dc:creator>
            <pubDate>Thu, 01 Oct 2026 13:19:29 GMT</pubDate>
            <atom:updated>2026-10-01T17:34:24.580Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/468/1*sARmQNzcof0z9hq4POET3g.jpeg" /></figure><p>After close to a decade as an engineer, I decided to take a sabbatical and enrolled at Seneca College, and moved from New Delhi, India to Toronto, Canada almost 15 years ago.</p><p>Several people I worked with found this pretty strange, after all going back to school after a solid run in the industry is not very common.</p><p>Recruiters were even more confused — they kept asking if the career break meant something had gone wrong, as if stepping back on purpose and getting pushed out were the same story.</p><p>A few former colleagues in New Delhi asked, half-joking, if I was having some kind of midlife crisis. I told them a midlife crisis usually involves a motorcycle, not a student ID card.</p><p>What none of them knew is that I wasn’t going back to school to learn a new stack or chase a new title. Communication had always been my quiet superpower — I just wanted to stop treating it like a lucky accident and start treating it like a skill worth actually training.</p><p>That’s the strange thing about a strength you’ve never formally studied. You can rely on it for a decade and still not be able to explain it to a single person, including yourself.</p><p>I’d love to hear — what’s a move in your own career that broke the “expected path,” and how did it turn out?</p><p>Connect with me on LinkedIn — two decades of DevOps and multi-cloud strategy, $3M in FinOps wins, and a track record of shipping through real fires, real deadlines, and real 2 AM PagerDuty alerts: <a href="https://www.linkedin.com/in/gurpreetsinghkambo?source=user_about----------------------3f9fab711112------------------------">linkedin.com/in/gurpreetsinghkambo</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=538eb1207e6f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>