<?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 Tech Trends Today on Medium]]></title>
        <description><![CDATA[Stories by Tech Trends Today on Medium]]></description>
        <link>https://medium.com/@sirtcp?source=rss-3299d8448ade------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*DAISpzv-dSq4i8bdRkc1NQ.jpeg</url>
            <title>Stories by Tech Trends Today on Medium</title>
            <link>https://medium.com/@sirtcp?source=rss-3299d8448ade------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 23:33:41 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@sirtcp/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[Troubleshooting Kubernetes Master Node Certificate Expiry and API Server Access Issues: A…]]></title>
            <link>https://medium.com/@sirtcp/troubleshooting-kubernetes-master-node-certificate-expiry-and-api-server-access-issues-a-8d265bdbbf33?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/8d265bdbbf33</guid>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[docker]]></category>
            <category><![CDATA[cloud-security]]></category>
            <category><![CDATA[linux]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Sat, 30 Aug 2025 10:25:57 GMT</pubDate>
            <atom:updated>2025-08-30T10:25:57.489Z</atom:updated>
            <content:encoded><![CDATA[<h3>Troubleshooting Kubernetes Master Node Certificate Expiry and API Server Access Issues: A Step-by-Step Guide</h3><p>Managing a Kubernetes cluster involves maintaining certificates, ensuring API server accessibility, and keeping nodes functional. Recently, I faced an issue where the kube-apiserver on my master node kept crashing due to expired certificates, followed by kubeconfig authentication errors. Here is the detailed step-by-step guide to troubleshoot and resolve these issues.</p><h3>Step 1: Identify the kube-apiserver Crash</h3><p>On the master node, I noticed the kube-apiserver pod was crashing repeatedly:</p><p>upon runing kubectl get pods</p><pre>journalctl -u kubelet -f | grep kube-apiserver</pre><p>Output:</p><pre>failed to &quot;StartContainer&quot; for &quot;kube-apiserver&quot; with CrashLoopBackOff</pre><p>I verified that there were no port conflicts on 6443:</p><pre>netstat -tulnp | grep 6443</pre><p>Output showed <strong>no conflicts</strong>, confirming that something else was causing the crash.</p><h3>Step 2: Check Kubernetes Certificates</h3><p>All Kubernetes components rely on certificates in /etc/kubernetes/pki. Listing the certificates:</p><pre>ls -l /etc/kubernetes/pki/</pre><p>To check if the apiserver certificate was expired:</p><pre>openssl x509 -in /etc/kubernetes/pki/apiserver.crt -noout -dates</pre><p>Output:</p><pre>notBefore=Jul  8 15:23:31 2024 GMT<br>notAfter=Jul  8 15:28:32 2025 GMT</pre><p>The certificate was <strong>already expired</strong>.</p><h3>Step 3: Check Expiration of All Kubernetes Certificates</h3><p>Kubernetes provides a handy command to check all certificates:</p><pre>kubeadm certs check-expiration</pre><p>Output:</p><pre>CERTIFICATE                EXPIRES                  RESIDUAL TIME<br>admin.conf                 Jul 08, 2025 15:28 UTC   &lt;invalid&gt;<br>apiserver                  Jul 08, 2025 15:28 UTC   &lt;invalid&gt;<br>...<br>CERTIFICATE AUTHORITY<br>ca                         Jul 06, 2034 15:28 UTC   8y</pre><p>All cluster certificates had expired, while the CA was still valid.</p><h3>Step 4: Renew Certificates</h3><p>To rene\w all certificates:</p><pre>kubeadm certs renew all</pre><p>After renewal, verify:</p><pre>kubeadm certs check-expiration</pre><p>The certificates now show valid residual time.</p><h3>Step 5: Fix API Server Credentials Issue</h3><p>After renewing certificates, kubectl returned:</p><pre>error: You must be logged in to the server (the server has asked for the client to provide credentials)</pre><h3>Diagnosis</h3><p>The existing admin.conf had the <strong>wrong API server URL</strong>:</p><pre>expected: https://192.168.1.75:6443, got: https://kmaster1:6443</pre><h3>Fix: Regenerate Admin Kubeconfig</h3><ol><li>Backup the existing kubeconfig:</li></ol><pre>mv /etc/kubernetes/admin.conf /etc/kubernetes/admin.conf.backup</pre><ol><li>Regenerate admin kubeconfig with the correct API server IP:</li></ol><pre>kubeadm init phase kubeconfig admin --apiserver-advertise-address=192.168.1.75</pre><p>Replace 192.168.1.75 with your master node’s actual IP.</p><ol><li>Export the new kubeconfig:</li></ol><pre>export KUBECONFIG=/etc/kubernetes/admin.conf</pre><p>Optional: Copy to home directory for convenience:</p><pre>mkdir -p $HOME/.kube<br>cp /etc/kubernetes/admin.conf $HOME/.kube/config<br>chown $(id -u):$(id -g) $HOME/.kube/config</pre><h3>Step 6: Verify Cluster Status</h3><p>Test access to the API server:</p><pre>kubectl get nodes<br>kubectl get pods -n kube-system</pre><p>Output should list all nodes and system pods without authentication errors.</p><h3>Step 7: Summary of Key Troubleshooting Steps</h3><ol><li>Checked kube-apiserver crash logs.</li><li>Verified no port conflicts for 6443.</li><li>Checked expiration of certificates.</li><li>Renewed expired Kubernetes certificates.</li><li>Fixed kubeconfig by regenerating it with the correct API server URL.</li><li>Verified cluster health using kubectl.</li></ol><h3>Conclusion</h3><p>Expired certificates and mismatched kubeconfig are common reasons for a crashing kube-apiserver and authentication errors. Following this step-by-step workflow ensures your cluster remains operational and secure.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8d265bdbbf33" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Why ZFS + Proxmox Is a Power Combo for Homelab and Enterprise Virtualization]]></title>
            <link>https://medium.com/@sirtcp/why-zfs-proxmox-is-a-power-combo-for-homelab-and-enterprise-virtualization-979f9e2f1a8e?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/979f9e2f1a8e</guid>
            <category><![CDATA[proxmox]]></category>
            <category><![CDATA[nas]]></category>
            <category><![CDATA[zfs]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[linux]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Fri, 23 May 2025 00:22:15 GMT</pubDate>
            <atom:updated>2025-05-23T00:22:15.305Z</atom:updated>
            <content:encoded><![CDATA[<p>If you’re running <strong>Proxmox VE</strong> for virtualization and not using <strong>ZFS</strong>, you’re missing out on one of the most powerful and flexible storage options available. ZFS (Zettabyte File System) isn’t just a filesystem — it’s a volume manager, data protector, snapshot tool, and performance tuner all in one.</p><p>When combined with <strong>Proxmox</strong>, ZFS unlocks features that go beyond what traditional filesystems can offer. Whether you’re building a resilient home lab or scaling a production hypervisor cluster, this duo delivers <strong>rock-solid performance, redundancy, and efficiency</strong>.</p><h3>🎯 Why Use ZFS with Proxmox?</h3><p>Proxmox VE has <strong>first-class support</strong> for ZFS. With it, you can:</p><ul><li>Create <strong>native ZFS-backed storage pools</strong> for VMs and containers</li><li>Take <strong>instant snapshots and rollbacks</strong> of running VMs</li><li>Use <strong>ZFS replication</strong> for fast disaster recovery</li><li>Set up <strong>RAID-Z or mirrors</strong> with no need for hardware RAID</li><li>Get <strong>integrity checks</strong> and <strong>self-healing</strong> for bit rot and silent corruption</li></ul><p>It’s like giving your virtualization stack a <strong>built-in SAN</strong> — but with open-source tools.</p><h3>🛠️ Setting Up ZFS in Proxmox</h3><p>If you’re installing Proxmox from scratch, just choose <strong>ZFS as your root filesystem</strong> during installation. But you can also create a ZFS pool later from the CLI.</p><h3>Create a ZFS Pool (RAID 1 Mirror Example):</h3><pre>zpool create zfs-storage mirror /dev/sdb /dev/sdc</pre><blockquote><em>This creates a mirrored pool called </em><em>zfs-storage.</em></blockquote><h3>Add ZFS as a Storage Option in Proxmox:</h3><ol><li>Go to <strong>Datacenter → Storage → Add → ZFS</strong></li><li>Give it an ID, like zfs-vmstore</li><li>Select the pool zfs-storage</li><li>Tick “Content” → Disk image, Container, etc</li><li>Save.</li></ol><p>Your VMs and containers can now live on this ZFS-backed pool.</p><h3>🧪 Snapshots &amp; Rollbacks: One Command Magic</h3><p>ZFS snapshots are <strong>instant and space-efficient</strong>, perfect for backing up or testing upgrades.</p><h3>Snapshot a VM disk (assume VM ID 101):</h3><pre>zfs snapshot zfs-storage/subvol-101-disk-0@pre-update</pre><p>Need to revert?</p><pre>zfs rollback zfs-storage/subvol-101-disk-0@pre-update</pre><p>Done. No downtime. No gigabytes copied.</p><h3>🔁 ZFS Replication = Built-In DR</h3><p>Want off-site or backup node replication?</p><p>Proxmox lets you set up <strong>automatic ZFS replication</strong>:</p><pre>Datacenter → Replication → Add</pre><p>Set source VM, target node, and schedule. It will <strong>send only changed blocks</strong>, making replication fast and bandwidth-friendly.</p><h3>💾 Compression and Dedup: Save Disk Space</h3><p>ZFS allows <strong>on-the-fly compression</strong>. Enable it:</p><pre>zfs set compression=lz4 zfs-storage</pre><p>This reduces VM disk usage with zero performance hit in most cases.</p><blockquote><em>Want deduplication? It’s available — but </em><strong><em>use with caution</em></strong><em>, as it requires a lot of RAM.</em></blockquote><h3>🧠 Monitoring and Health Checks</h3><p>ZFS scrubs check for corruption and auto-heal it (if mirrored or RAID-Z):</p><pre>zpool scrub zfs-storage<br>zpool status</pre><p>Run this <strong>weekly or monthly</strong> to keep your pool healthy.</p><h3>🔐 Bonus: ZFS Send + Encryption</h3><p>ZFS also supports <strong>encryption</strong> and <strong>send/receive</strong> workflows:</p><pre>zfs send zfs-storage/subvol-101-disk-0@snapshot | ssh backup-node zfs recv backup-pool/vm-backup</pre><p>You now have <strong>encrypted snapshots</strong> replicated securely to another server.</p><h3>🎤 Final Thoughts</h3><p>Pairing <strong>ZFS with Proxmox</strong> gives you a flexible, battle-tested platform with enterprise-class storage capabilities — <strong>without enterprise pricing</strong>.</p><p>Whether you’re backing up VMs, running a high-availability cluster, or simply tired of worrying about disk corruption, <strong>ZFS + Proxmox is a match made in sysadmin heaven</strong>.</p><h3>🏷️ Tags</h3><pre>#Proxmox #ZFS #Virtualization #SysAdmin #Homelab #OpenSource #Linux #Storage #Snapshots #Replication #DevOps</pre><p>Would you like a <strong>companion cheat sheet</strong> PDF for Proxmox-ZFS workflows, or a comparison between <strong>ZFS vs Ceph for Proxmox</strong>?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=979f9e2f1a8e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ ZFS: Why This Filesystem is a Game-Changer for SysAdmins]]></title>
            <link>https://medium.com/@sirtcp/zfs-why-this-filesystem-is-a-game-changer-for-sysadmins-82466ee8398b?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/82466ee8398b</guid>
            <category><![CDATA[storage]]></category>
            <category><![CDATA[nas]]></category>
            <category><![CDATA[zfs]]></category>
            <category><![CDATA[sysadmin]]></category>
            <category><![CDATA[linux]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Thu, 22 May 2025 21:20:59 GMT</pubDate>
            <atom:updated>2025-05-22T21:20:59.988Z</atom:updated>
            <content:encoded><![CDATA[<p>When it comes to managing storage like a pro, <strong>ZFS</strong> is the tool that makes you feel like a rock star. Originally developed by Sun Microsystems, ZFS (Zettabyte File System) is more than just a filesystem — it’s a <strong>storage management platform</strong> that combines a volume manager, filesystem, and RAID into one cohesive system.</p><p>Here’s why ZFS deserves the hype — and how you can start using it like a legend.</p><h3>🎯 What Makes ZFS Special?</h3><p>Most filesystems do <em>one job</em> — they store your data. ZFS does that, and then <em>10 more</em>. It brings together the best of volume management, data integrity, and simplified administration.</p><h3>Key Features That Shine:</h3><ul><li><strong>Copy-on-Write (CoW)</strong>: No data is overwritten — every change creates a new snapshot, ensuring data consistency.</li><li><strong>Data Integrity</strong>: ZFS uses checksums to detect and fix silent data corruption. No more bit rot nightmares.</li><li><strong>Snapshots and Clones</strong>: Create instant snapshots without taking your system offline. Even better — create writable clones from them!</li><li><strong>RAID-Z</strong>: Say goodbye to traditional RAID headaches. ZFS handles redundancy with built-in RAID-Z (levels 1, 2, and 3).</li><li><strong>Compression</strong>: Enable transparent compression with a single command.</li><li><strong>Self-Healing</strong>: With mirrored or RAID-Z setups, ZFS can automatically fix data errors.</li></ul><h3>⚙️ Getting Started with ZFS on Linux</h3><p>ZFS is not just for Solaris anymore. It’s widely available on <strong>Linux</strong> (via ZFS on Linux), <strong>FreeBSD</strong>, and even <strong>macOS (via MacZFS)</strong>.</p><h3>🛠️ Installation (Ubuntu/Debian</h3><pre>sudo apt update<br>sudo apt install zfsutils-linux</pre><h3>🎸 Create a Pool (Think of it like a virtual RAID)</h3><pre>sudo zpool create rockpool /dev/sdb /dev/sdc</pre><blockquote><em>Want RAID-Z? Use:<br> </em><em>sudo zpool create rockpool raidz /dev/sdb /dev/sdc /dev/sdd</em></blockquote><h3>🧠 Create a Filesystem</h3><pre>sudo zfs create rockpool/mydata</pre><p>ZFS filesystems are <em>automatically mounted</em> at /rockpool/mydata. You don’t need /etc/fstab anymore.</p><h3>📸 Snapshots &amp; Clones Like a Boss</h3><p>Create a snapshot:</p><pre>sudo zfs snapshot rockpool/mydata@before-upgrade</pre><p>Roll back if something breaks:</p><pre>sudo zfs rollback rockpool/mydata@before-upgrade</pre><p>Create a clone for testing:</p><pre>sudo zfs clone rockpool/mydata@before-upgrade rockpool/dev-clone</pre><h3>💾 Transparent Compression</h3><p>ZFS supports several compression algorithms. Enable compression like this:</p><pre>sudo zfs set compression=lz4 rockpool/mydata</pre><p>Check compression stats:</p><pre>sudo zfs get compressratio,compression rockpool/mydata</pre><h3>🚨 Monitor and Heal</h3><p>ZFS constantly checks for issues. Run a scrub to detect and fix problems:</p><pre>sudo zpool scrub rockpool</pre><p>Check pool health:</p><pre>sudo zpool status</pre><h3>🧪 Bonus: Deduplication</h3><p>ZFS can deduplicate blocks of data — if the same data is written multiple times, it stores just one copy.</p><pre>sudo zfs set dedup=on rockpool/mydata</pre><p>⚠️ <strong>Warning</strong>: Deduplication is RAM-intensive. Only enable it when absolutely needed.</p><h3>🧘‍♂️ Why SysAdmins Love ZFS</h3><p>ZFS simplifies complex storage setups. Whether you’re managing a home NAS, an enterprise database server, or a cloud backup system, ZFS gives you peace of mind with features that traditionally required expensive SAN appliances.</p><p>From silent data corruption detection to real-time rollback, ZFS is the toolkit that turns a cautious sysadmin into a confident one.</p><h3>🎤 Final Thoughts</h3><p>ZFS isn’t just another filesystem. It’s a <strong>data guardian</strong>, <strong>performance tuner</strong>, and <strong>storage orchestrator</strong> — all rolled into one. Whether you’re running your infrastructure on bare metal or virtual machines, ZFS is your backstage pass to reliable and scalable storage.</p><p>So go ahead — <strong>create a pool</strong>, take a snapshot, and feel the power. Because in the world of filesystems, <strong>ZFS is the rock star.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=82466ee8398b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[10 Underrated Linux System Administration Tips & Tricks Every New Admin Should Know]]></title>
            <link>https://medium.com/@sirtcp/10-underrated-linux-system-administration-tips-tricks-every-new-admin-should-know-bb3c6c399821?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/bb3c6c399821</guid>
            <category><![CDATA[linux]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[network-security]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[sysadmin]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Thu, 22 May 2025 20:41:54 GMT</pubDate>
            <atom:updated>2025-05-22T20:41:54.019Z</atom:updated>
            <content:encoded><![CDATA[<p>As a new Linux system administrator, you’re likely familiar with the basics — top, df, systemctl, journalctl, and others. But some truly powerful and lesser-known tools and techniques can make your day-to-day work significantly easier, more efficient, and even more secure.</p><p>Here are <strong>10 underrated Linux system administration tips and tricks</strong> that aren’t always part of the standard playbook but can help you stand out as a smart, efficient sysadmin.</p><h3>1. 🔁 Use watch to Monitor Changes in Real Time</h3><p>Instead of running the same command over and over again, use watch to monitor the output every few seconds.</p><pre>watch -d -n 2 df -h</pre><ul><li>-d: Highlights the differences.</li><li>-n 2: Runs every 2 seconds.</li></ul><p>Great for watching disk space, load, or service status.</p><h3>2. 📁 Use ncdu for Disk Usage (Better Than du)</h3><p>ncdu (NCurses Disk Usage) is a simple but powerful alternative to du for investigating disk usage.</p><pre>sudo apt install ncdu    # Debian/Ubuntu<br>sudo ncdu /</pre><p>It presents an interactive, scrollable UI showing where disk space is used. Much quicker than parsing du -sh.</p><h3>3. ⛑ Use ss Instead of netstat</h3><p>ss is faster and more accurate than the older netstat for checking open ports and connections.</p><pre>ss -tuln</pre><ul><li>-t: TCP</li><li>-u: UDP</li><li>-l: Listening sockets</li><li>-n: Show numeric addresses</li></ul><p>Bonus:</p><pre>ss -tp   # Shows which process is using which port</pre><h3>4. 📜 Quick Audit With last and who</h3><p>To check who has logged into the system and when:</p><pre>last<br>who</pre><p>You can even combine with grep for filtering:</p><pre>last | grep &quot;ssh&quot;</pre><h3>5. 🛑 Limit Dangerous rm with Safer Defaults</h3><p>Accidental rm -rf / situations are every admin’s nightmare. Add this to your .bashrc or /etc/profile.d/safety.sh:</p><pre>alias rm=&#39;rm -i&#39;<br>alias cp=&#39;cp -i&#39;<br>alias mv=&#39;mv -i&#39;</pre><p>This makes file removal interactive by default.</p><h3>6. 🔐 Check for SUID/SGID Files (Security Audit)</h3><p>To quickly check for potentially risky binaries with elevated permissions:</p><pre>find / -perm /6000 -type f -exec ls -ld {} \; 2&gt;/dev/null</pre><p>Files with SUID/SGID can be exploited if misconfigured.</p><h3>7. 🧠 ps auxf – Visual Process Tree</h3><p>Better than plain ps aux, this gives you a hierarchical view of processes:</p><pre>ps auxf</pre><p>Use it with less for paging:</p><pre>ps auxf | less</pre><p>It helps visualize parent/child relationships between processes.</p><h3>8. 🕵️‍♂️ Find Recently Modified Files</h3><p>Troubleshooting something changed? Find recently modified files:</p><pre>find /etc -type f -mtime -1</pre><ul><li>-mtime -1: Modified in the last 24 hours.</li><li>You can adjust for minutes with -mmin.</li></ul><p>Bonus for auditing cron jobs or changes:</p><pre>find /var/spool/cron -type f -mtime -1</pre><h3>9. 🧹 Clean Up Systemd Journal Logs</h3><p>Systemd’s journal can eat disk over time. Clean it up with:</p><pre>sudo journalctl --vacuum-time=7d</pre><p>Or by size:</p><pre>sudo journalctl --vacuum-size=500M</pre><p>Add this to a monthly cron job to prevent log bloat.</p><h3>10. 📡 Use curl ifconfig.me for External IP (No GUI Tools)</h3><p>Need to know your server’s public IP quickly?</p><pre>curl ifconfig.me</pre><p>Or, more verbose:</p><pre>curl -s ifconfig.me/all</pre><p>This is handy in cloud setups, firewall rules, or NAT debugging.</p><h3>🧩 Bonus: alias alert=&#39;notify-send --urgency=low -i terminal &quot;Command completed&quot;&#39;</h3><p>If you’re running a long task and want a notification when it finishes:</p><pre>sleep 10; alert</pre><p>Pair it with commands:</p><pre>long_command; alert</pre><p>This only works on systems with a GUI and notify-send.</p><h3>Final Thoughts</h3><p>These lesser-known but highly effective tips can significantly boost your productivity and insight as a Linux system administrator. Knowing how to get deeper visibility into system behavior or how to protect against small but dangerous mistakes can make you a more reliable and trusted admin.</p><p>If you’re starting your journey, try incorporating 2–3 of these techniques into your daily routine. Over time, they’ll become second nature and separate you from the average crowd.</p><p>Want a PDF or need these adapted to a particular distro like CentOS or Alpine? Let me know!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bb3c6c399821" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[️ How to Install Proxmox VE with Software RAID Using MDADM (The Easy Way)]]></title>
            <link>https://medium.com/@sirtcp/%EF%B8%8F-how-to-install-proxmox-ve-with-software-raid-using-mdadm-the-easy-way-ad7b050fec47?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/ad7b050fec47</guid>
            <category><![CDATA[proxmox]]></category>
            <category><![CDATA[virtualization]]></category>
            <category><![CDATA[raid]]></category>
            <category><![CDATA[linux]]></category>
            <category><![CDATA[self-hosting]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Thu, 22 May 2025 00:01:07 GMT</pubDate>
            <atom:updated>2025-05-22T00:01:07.334Z</atom:updated>
            <content:encoded><![CDATA[<p>If you’re planning to build a Proxmox VE (Virtual Environment) server but don’t have access to a hardware RAID controller, don’t worry — you can still achieve redundancy and disk reliability using <strong>MDADM</strong>, the Linux software RAID utility.</p><p>In this article, I’ll walk you through a clean and reliable way to install Proxmox with <strong>MDADM-based software RAID</strong>, perfect for home labs, low-budget servers, or testing environments.</p><h3>⚙️ What You’ll Need</h3><ul><li>Two or more physical drives of equal siz</li><li>A USB stick with a <strong>Debian 12 (Bookworm)</strong> netinst ISO</li><li>Internet connectivity during install</li><li>Basic familiarity with the Linux shell</li><li>Time: ~45–60 minutes for full install</li></ul><h3>❓ Why Not Just Use ZFS?</h3><p>Proxmox does support <strong>ZFS RAID</strong> out of the box during installation, and it’s a great choice. But there are reasons why you might prefer <strong>MDADM</strong>:</p><ul><li>Lighter resource usage (ZFS is RAM hungry)</li><li>Broader compatibility with traditional Linux tools</li><li>You want to stick to EXT4 or XFS file systems</li><li>Better compatibility with legacy systems</li></ul><h3>🛠️ Step-by-Step: Install Proxmox VE with MDADM RAID</h3><h3>🔹 Step 1: Install Debian 12 (Bookworm)</h3><p>Download the Debian <strong>netinst ISO</strong> from <a href="https://www.debian.org/distrib/">https://www.debian.org/distrib/</a> and boot from it.</p><p>During installation:</p><ul><li>Choose <strong>Expert Install</strong> (for better control)</li><li>Partition <strong>only one disk</strong> at first</li><li>Set hostname, root password, user account</li><li>Choose <strong>“Standard system utilities”</strong> only (uncheck desktop environments)</li></ul><p>Complete the base installation and <strong>reboot into Debian</strong>.</p><h3>🔹 Step 2: Install MDADM and Create RAID</h3><p>After the first boot:</p><pre>sudo apt update<br>sudo apt install mdadm</pre><p>Now let’s create a RAID1 mirror. Assuming your disks are /dev/sda and /dev/sdb, and you only used /dev/sda during install:</p><p><strong>Create RAID1 array:</strong></p><pre>sudo mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sda /dev/sdb</pre><p>Initialize and save the config:</p><pre>sudo mdadm --detail --scan | sudo tee -a /etc/mdadm/mdadm.conf<br>sudo update-initramfs -u</pre><h3>🔹 Step 3: Partition and Format the RAID</h3><p>Partition /dev/md0 as you wish:</p><pre>sudo fdisk /dev/md0</pre><p>Create one root partition or separate /boot, /, and /var if desired. Then format:</p><pre>sudo mkfs.ext4 /dev/md0p1</pre><p>Mount it:</p><pre>sudo mount /dev/md0p1 /mnt</pre><h3>🔹 Step 4: Install Proxmox VE</h3><p>Follow these commands carefully:</p><pre>echo &quot;deb http://download.proxmox.com/debian/pve bookworm pve-no-subscription&quot; | sudo tee /etc/apt/sources.list.d/pve-install.list<br>wget https://enterprise.proxmox.com/debian/proxmox-release-bookworm.gpg -O- | sudo gpg --dearmor -o /etc/apt/trusted.gpg.d/proxmox-release-bookworm.gpg<br>sudo apt update<br>sudo apt install proxmox-ve postfix open-iscsi -y</pre><p>Disable enterprise repo (if you don’t have a subscription):</p><pre>sudo sed -i &quot;s/^deb/#deb/&quot; /etc/apt/sources.list.d/pve-enterprise.list</pre><p>Reboot the system:</p><pre>sudo reboot</pre><p>Proxmox should now be installed and accessible via:</p><pre>https://your-server-ip:8006</pre><h3>💡 Bonus: Set up GRUB for Boot Redundancy</h3><p>If you want both drives to be bootable:</p><pre>sudo grub-install /dev/sdb</pre><p>Then update GRUB:</p><pre>sudo update-grub</pre><p>Now if /dev/sda fails, the system can still boot from /dev/sdb.</p><h3>✅ Final Notes</h3><ul><li>RAID is not a backup — always sync offsite or to another storage</li><li>Monitor the RAID health regularly with mdadm --detail /dev/md0</li><li>Consider adding <strong>email alerts</strong> for disk failures using mdadm configs</li></ul><h3>📌 Conclusion</h3><p>Using <strong>MDADM</strong> with Proxmox gives you more flexibility and works great when you don’t have hardware RAID or want full software-level control over your disks. It’s stable, performant, and more than enough for small-to-mid-scale virtualization workloads.</p><p>Whether you’re building a home lab or deploying on budget hardware, this method gives you both redundancy and freedom.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ad7b050fec47" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ 10 Essential NGINX Tips for Web Admins (With Commands)]]></title>
            <link>https://medium.com/@sirtcp/10-essential-nginx-tips-for-web-admins-with-commands-44f6f16c1ddc?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/44f6f16c1ddc</guid>
            <category><![CDATA[nginx]]></category>
            <category><![CDATA[reverse-proxy]]></category>
            <category><![CDATA[web-server]]></category>
            <category><![CDATA[linux]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Wed, 21 May 2025 18:59:09 GMT</pubDate>
            <atom:updated>2025-05-21T18:59:09.268Z</atom:updated>
            <content:encoded><![CDATA[<p>Whether you’re running a hobby blog or managing enterprise apps, <strong>NGINX</strong> remains a powerful tool for serving web traffic efficiently. Here are 10 practical tips — complete with commands — that can enhance your NGINX workflow, security, and performance.</p><h3>1. 🔍 Test Configuration Before Reload</h3><p>Avoid breaking your live server with bad config.</p><p>If valid, reload safely:</p><pre>sudo systemctl reload nginx</pre><h3>2. 🔁 Enable Gzip Compression</h3><p>Speed up load times with compressed assets.</p><p>In your NGINX config (/etc/nginx/nginx.conf):</p><pre>gzip on;<br>gzip_types text/plain text/css application/json application/javascript text/xml application/xml application/xml+rss text/javascript;</pre><h3>3. 🔐 Set Up Basic Auth for Admin Panels</h3><p>Protect sensitive areas with a username and password.</p><ol><li>Create password file:</li></ol><pre>sudo apt install apache2-utils<br>htpasswd -c /etc/nginx/.htpasswd adminuser</pre><ol><li>Add this to your server block:</li></ol><pre>location /admin {<br>    auth_basic &quot;Restricted&quot;;<br>    auth_basic_user_file /etc/nginx/.htpasswd;<br>}</pre><h3>4. ⚡ Serve Static Files Efficiently</h3><p>Let NGINX handle images, CSS, and JS directly.</p><pre>location /static/ {<br>    alias /var/www/yourapp/static/;<br>    expires 30d;<br>}</pre><h3>5. 🔄 Set Up Reverse Proxy</h3><p>Forward requests to another service (e.g., Node.js or Flask app).</p><pre>location / {<br>    proxy_pass http://127.0.0.1:3000;<br>    proxy_set_header Host $host;<br>    proxy_set_header X-Real-IP $remote_addr;<br>}</pre><h3>6. 🛑 Block User Agents (e.g., bots)</h3><p>Stop scrapers or malicious traffic with bad user agents.</p><pre>if ($http_user_agent ~* (crawler|bot|spider)) {<br>    return 403;<br>}</pre><h3>7. 🕵️‍♂️ Enable Access and Error Logging</h3><p>Useful for debugging and audits.</p><pre>access_log /var/log/nginx/access.log;<br>error_log /var/log/nginx/error.log warn;</pre><p>You can also <strong>turn off</strong> access logging for specific paths:</p><pre>location /healthcheck {<br>    access_log off;<br>}</pre><h3>8. 📈 Limit Request Rate (Basic DoS Protection)</h3><p>Throttle excessive requests:</p><pre>limit_req_zone $binary_remote_addr zone=api_limit:10m rate=10r/s;</pre><pre>location /api/ {<br>    limit_req zone=api_limit burst=20;<br>}</pre><h3>9. 🌐 Redirect HTTP to HTTPS</h3><p>Force secure traffic:</p><pre>server {<br>    listen 80;<br>    server_name example.com www.example.com;<br>    return 301 https://$host$request_uri;<br>}</pre><h3>10. 🔄 Auto-Restart NGINX on Crash (Optional)</h3><p>Use systemd to restart on failure:</p><pre>sudo systemctl edit nginx</pre><p>Add this under [Service]:</p><pre>Restart=always<br>RestartSec=5</pre><h3>💡 Bonus: Reload NGINX Without Dropping Connections</h3><p>For zero-downtime reloads (built-in):</p><pre>sudo nginx -s reload</pre><p>Or gracefully via systemd:</p><pre>sudo systemctl reload nginx</pre><h3>✅ Final Thoughts</h3><p>NGINX is a flexible beast — a simple web server, a load balancer, a reverse proxy, and even a WAF with the right modules. With these tips in your toolkit, you’re better prepared to run a secure, high-performance server environment.</p><p>📌 <strong>Bookmark this</strong> or fork it into your own admin wiki — and if you’d like more tips on NGINX tuning, SSL, or automation with Ansible, drop a comment or DM!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=44f6f16c1ddc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Copilot vs ChatGPT: Are Your Secrets Safe With AI Coding Tools?]]></title>
            <link>https://medium.com/@sirtcp/copilot-vs-chatgpt-are-your-secrets-safe-with-ai-coding-tools-f2d503de7209?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/f2d503de7209</guid>
            <category><![CDATA[ai-programming]]></category>
            <category><![CDATA[copilot-programming]]></category>
            <category><![CDATA[github]]></category>
            <category><![CDATA[chatgpt]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Wed, 21 May 2025 18:45:27 GMT</pubDate>
            <atom:updated>2025-05-28T22:48:23.589Z</atom:updated>
            <content:encoded><![CDATA[<p>Artificial intelligence tools like <strong>GitHub Copilot</strong> and <strong>ChatGPT</strong> have become indispensable for developers. They can write code, generate boilerplates, and even explain complex concepts in seconds. But as these tools grow in popularity, so do the concerns around <strong>data privacy</strong> — especially when it comes to proprietary code, API keys, and business logic.</p><p>You may have heard the warning:</p><blockquote>“Don’t paste your secrets into ChatGPT!”</blockquote><p>Yet many of us use <strong>GitHub Copilot</strong>, which is powered by OpenAI’s models — the same technology that drives ChatGPT. So why is <strong>Copilot considered secure</strong>, while <strong>ChatGPT raises red flags</strong> for sensitive code?</p><p>Let’s break it down.</p><h3>🤖 What Are Copilot and ChatGPT?</h3><p>ToolDescription<strong>GitHub Copilot</strong>AI coding assistant integrated into IDEs like VS Code and JetBrains. Provides code suggestions and autocompletion.<strong>ChatGPT</strong>General-purpose conversational assistant by OpenAI. Can also write and explain code, among many other tasks.</p><p>While they serve different interfaces and workflows, both tools rely on <strong>OpenAI’s large language models</strong> — like Codex or GPT-4.</p><h3>🔐 The Privacy Concern</h3><p>The key worry with using AI tools for coding is <strong>data leakage</strong>. If you paste proprietary source code, API keys, or internal logic into these tools, could that data be stored or used to train future models?</p><p>With ChatGPT, the answer is <strong>yes</strong> — unless you take specific steps.</p><h3>🧠 How GitHub Copilot and ChatGPT Handle Your Data</h3><p><strong>General comparision :</strong></p><pre>+----------------------------+-------------------------------+----------------------------------------------+<br>|         Feature          |        GitHub Copilot         |           ChatGPT (Free/Plus)                |<br>+----------------------------+-------------------------------+----------------------------------------------+<br>| Provider                 | GitHub (Microsoft)            | OpenAI                                       |<br>| Use Case                 | Coding assistant in IDE       | General-purpose conversational assistant     |<br>| Model                    | OpenAI Codex / GPT            | GPT-4 / GPT-4o                               |<br>| Code Sent for Training?  | ❌ Not if on Copilot Business | ✅ Yes, by default for Free/Plus users       |<br>| Option to Opt Out?       | ✅ Yes (telemetry &amp; training) | ✅ Yes (via settings)                        |<br>| Enterprise Privacy       | ✅ Strong protections         | ✅ Only with ChatGPT Team/Enterprise         |<br>+----------------------------+-------------------------------+----------------------------------------------+</pre><p><strong>DATA Privacy :</strong></p><pre>+------------------------+-------------------------------+----------------------------------------------+<br>|        Aspect          |     Microsoft Copilot         |                  ChatGPT                     |<br>+------------------------+-------------------------------+----------------------------------------------+<br>| Data Processing        | Localized within M365         | Cloud-based on OpenAI servers                |<br>| Data Retention         | Enterprise-controlled,        | Stores history by default,                  |<br>|                        | zero retention mode available | opt-out available                            |<br>| Compliance             | GDPR, CCPA, ISO/IEC 27001     | General privacy settings,                    |<br>|                        | compliant                     | no enterprise controls                       |<br>| Data Use for Training  | Does not use user data        | Uses data unless opted out                   |<br>|                        | for training                  |                                              |<br>| Encryption             | End-to-end encryption         | No explicit end-to-end encryption            |<br>|                        | within enterprise             |                                              |<br>| Suitability            | Regulated industries,         | General-purpose, low-sensitivity tasks       |<br>|                        | sensitive data                |                                              |<br>+------------------------+-------------------------------+----------------------------------------------+</pre><h3>🔒 Why Copilot for Business Is Safer</h3><p>GitHub Copilot <strong>for Business</strong> and <strong>Enterprise</strong> users benefit from:</p><ul><li><strong>No retention</strong> of your prompts or code for training purposes.</li><li><strong>Strict isolation</strong> of enterprise data.</li><li><strong>No secret logging</strong>, even in telemetry (if opted out).</li><li>Local integration with your IDE, so minimal context is sent to OpenAI’s servers.</li></ul><p>Compare that to <strong>ChatGPT Free or Plus</strong>, where your conversations <strong>can</strong> be used to improve the model unless you explicitly disable history and training.</p><h3>⚙️ How to Protect Your Code in ChatGPT</h3><p>If you’re using ChatGPT and still want to experiment safely:</p><ol><li><strong>Disable history and training</strong><br> → Go to Settings → Data Controls → Turn off “Chat history &amp; training”.</li><li><strong>Never paste secrets, keys, or proprietary data</strong><br> → Assume anything you enter is logged unless you’re using ChatGPT <strong>Enterprise</strong>.</li><li><strong>Use ChatGPT Team or Enterprise</strong> for secure usage<br> → These plans guarantee <strong>no training or retention</strong> of your data.</li></ol><h3>✅ Best Practices</h3><ul><li>For <strong>casual or public questions</strong>, ChatGPT Free/Plus is fine.</li><li>For <strong>private codebases or production-level work</strong>:</li><li>Use <strong>GitHub Copilot for Business</strong></li><li>Or use <strong>ChatGPT Enterprise/Team</strong></li><li>Always <strong>treat AI tools like third-party contractors</strong> — don’t overshare unless you’re on a secure, compliant platform.</li></ul><h3>💡 Final Thoughts</h3><p>AI tools like GitHub Copilot and ChatGPT are revolutionizing how we write code — but privacy and security <strong>must come first</strong>. While both use powerful OpenAI models, the difference is in <strong>how your data is handled</strong>.</p><p>If you’re a developer working on sensitive projects, don’t risk leaking secrets. Choose tools that align with your data protection needs, and always review their privacy documentation.</p><p>✍️ <em>Did you find this helpful? Follow me for more practical AI tips for developers.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f2d503de7209" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Troubleshooting Azure AD Connect Sync Failures: The Case of the “Unrecognized GUID Format”]]></title>
            <link>https://medium.com/@sirtcp/troubleshooting-azure-ad-connect-sync-failures-the-case-of-the-unrecognized-guid-format-20af113d8c10?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/20af113d8c10</guid>
            <category><![CDATA[active-directory]]></category>
            <category><![CDATA[ad-connect]]></category>
            <category><![CDATA[azure-active-directory]]></category>
            <category><![CDATA[windows]]></category>
            <category><![CDATA[async]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Tue, 20 May 2025 20:13:41 GMT</pubDate>
            <atom:updated>2025-05-20T20:13:41.812Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Azure AD Connect</strong> is a powerful tool that bridges your on-premises Active Directory with Azure Active Directory. It enables hybrid identity setups, single sign-on, and seamless user provisioning across environments.</p><p>But what happens when synchronization stops working, and all you get is a vague error like:</p><blockquote><em>sync-generic-failure or<br> </em><em>unrecognized GUID format</em></blockquote><p>If you’re here, you’ve probably been there — staring at logs inside the <strong>Synchronization Service Manager</strong>, wondering why updates are stuck, and scratching your head at a device object with a long GUID.</p><p>Here’s how to figure it out — and fix it.</p><h3>🛠️ The Problem</h3><p>In one case I investigated, a sync error was being caused by an object in Active Directory with a distinguished name like:</p><pre>CN=82467a34-xxxx-xxxx-xxxx-xxxxxxxxxxxx,CN=RegisteredDevices,DC=example,DC=local</pre><p>This was a <strong>device object</strong> created for hybrid Azure AD join, and it had an attribute (msDS-KeyCredentialLink) that was causing the sync to fail. The exact error in Azure AD Connect?</p><blockquote><em>❌ </em>“Unrecognized GUID format”</blockquote><p>This generic error usually appears under <strong>Connectors → Export Errors</strong> in the Sync Service Manager.</p><h3>🔍 Root Cause: Corrupted Device Attribute</h3><p>The attribute msDS-KeyCredentialLink is used by Windows Hello for Business and contains cryptographic keys. When this attribute becomes malformed, Azure AD Connect can’t parse it, causing sync failures — and unfortunately, the logs don’t always point directly to this attribute.</p><p>To identify the culprit:</p><pre>Get-ADObject -LDAPFilter &quot;(ObjectClass=msDS-Device)&quot; -Properties msDS-KeyCredentialLink | <br>Where-Object { $_.&#39;msDS-KeyCredentialLink&#39;.Count -gt 0 }</pre><p>Look for device objects with overly large or oddly formatted values in that attribute.</p><h3>✅ The Fix: Remove or Clean the Faulty Object</h3><p>If the object is no longer needed (e.g., a test device, stale join), the safest fix is to remove it entirely:</p><pre>Remove-ADObject -Identity &quot;CN=GUID,CN=RegisteredDevices,DC=example,DC=local&quot; -Confirm:$true</pre><p>Alternatively, you can clear just the msDS-KeyCredentialLink attribute:</p><pre>Set-ADObject -Identity &quot;CN=GUID,CN=RegisteredDevices,DC=example,DC=local&quot; `<br>  -Remove @{ &quot;msDS-KeyCredentialLink&quot; = $true }</pre><p>After that, trigger a delta sync:</p><pre>Start-ADSyncSyncCycle -PolicyType Delta</pre><h3>🔄 Best Practices to Avoid Future Sync Issues</h3><ul><li>✅ Regularly audit RegisteredDevices for stale or orphaned entries</li><li>✅ Avoid unnecessary device joins for test systems</li><li>✅ Monitor msDS-KeyCredentialLink attributes for unusually large values</li><li>✅ Keep Azure AD Connect updated — newer builds handle key parsing better</li></ul><h3>🚀 Final Thoughts</h3><p>Azure AD Connect is incredibly reliable — but even powerful systems can choke on malformed data. If you’re dealing with a sync-generic-failure and errors about GUIDs, don’t panic. It&#39;s usually a rogue device object with a messy key attribute.</p><p>With a bit of PowerShell sleuthing and cleanup, you’ll be back in sync in no time.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=20af113d8c10" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Setting Email Sending Limits in Postfix Using Postfwd2]]></title>
            <link>https://medium.com/@sirtcp/setting-email-sending-limits-in-postfix-using-postfwd2-653e4b8d6361?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/653e4b8d6361</guid>
            <category><![CDATA[postfix]]></category>
            <category><![CDATA[postfwd]]></category>
            <category><![CDATA[postfix-mail-server]]></category>
            <category><![CDATA[email-notifications]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Fri, 09 May 2025 23:36:05 GMT</pubDate>
            <atom:updated>2025-05-09T23:37:27.817Z</atom:updated>
            <content:encoded><![CDATA[<p>Postfwd2 is a powerful policy daemon for Postfix, enabling complex access controls, rate limiting, and anti-spam measures using flexible rule sets. This guide walks you through the entire process-from installation to advanced troubleshooting.</p><h3>Table of Contents</h3><ul><li>Introduction</li><li>Installation</li><li>Integration with Postfix</li><li>Writing Rules</li><li>Testing and Validation</li><li>Running and Managing the Service</li><li>Troubleshooting Common Issues</li><li>Best Practices and Tips</li><li>Conclusion</li></ul><h3>Introduction</h3><p><strong>What is Postfwd2?</strong><br> Postfwd2 is a Perl-based policy daemon that leverages the Postfix policy delegation protocol to enforce complex mail flow restrictions before messages are accepted. It can block, rate-limit, or score messages based on various SMTP parameters, such as sender, recipient, client IP, and more<a href="https://github.com/postfwd/postfwd">1</a><a href="https://manpages.ubuntu.com/manpages/focal/man1/postfwd2.1.html">2</a><a href="http://postfwd.org/DEVEL/postfwd2-0.18RC1.html">3</a>.</p><h3>Installation</h3><h3>On Debian/Ubuntu</h3><pre>sudo apt-get update<br>sudo apt-get install postfwd</pre><p>This will install postfwd2 and its dependencies from your distribution’s repositories.</p><h3>Manual Installation</h3><ol><li>Clone the repository:</li></ol><ul><li>git clone <a href="https://github.com/postfwd/postfwd.git">https://github.com/postfwd/postfwd.git</a></li></ul><ol><li>Copy the daemon to your path:</li></ol><ul><li>sudo cp ./postfwd/sbin/postfwd2 /usr/sbin/</li></ul><ol><li>Install Perl dependencies:</li></ol><ul><li>sudo apt-get install build-essential sudo cpan Net::Server::Daemonize Net::Server::Multiplex Net::Server::PreFork Net::DNS IO::Multiplex</li></ul><ol><li>(You may also need gcc, make, and other build tools)<a href="https://github.com/postfwd/postfwd">1</a><a href="https://forum.efa-project.org/viewtopic.php?t=1461">5</a>.</li></ol><h3>Integration with Postfix</h3><h3>1. Configure Postfwd2 to Listen for Policy Requests</h3><p>By default, Postfwd2 listens on 127.0.0.1:10040. You can customize this with the -i (interface) and -p (port) options.</p><h3>2. Edit Postfix Main Configuration</h3><p>Open /etc/postfix/main.cf and add or adjust:</p><pre>smtpd_recipient_restrictions =<br>    check_policy_service inet:127.0.0.1:10040,<br>    permit_mynetworks,<br>    reject_unauth_destination</pre><p><em>Place the </em><em>check_policy_service before </em><em>permit_mynetworks if you want Postfwd2 to process all connections, including trusted networks.</em></p><p>After editing, reload Postfix:</p><pre>sudo systemctl reload postfix</pre><h3>Writing Rules</h3><h3>Rule Syntax</h3><ul><li>Each rule is a single line, with fields separated by semicolons (;).</li><li><strong>Format:</strong></li><li>[item1=value; item2=value; ...] action=&lt;result&gt;</li><li>For rate limiting in postfwd2:</li><li>id=LIMIT_IP; client_address=192.168.0.10,192.168.0.11; action=rate($$client_address/2/86400/REJECT You have exceeded your daily quota of 2 emails.);</li><li>This limits the listed IPs to 2 emails per 24 hours.</li></ul><h3>Common Actions</h3><ul><li>REJECT &lt;reason&gt;: Reject the message.</li><li>DUNNO: No decision, continue with next restriction.</li><li>rate(...): Apply rate limiting.</li></ul><p><strong>Example: Block a specific sender</strong></p><pre>sender==spammer@example.com; action=REJECT Blocked sender;</pre><p><strong>Example: Rate limit by sender</strong></p><pre>action=rate($$sender/100/3600/REJECT Too many emails this hour.);</pre><h3>Testing and Validation</h3><h3>Check Rule Syntax</h3><pre>postfwd2 -C -f /etc/postfix/postfwd.cf</pre><p>This will validate your rules file and print any errors or warnings.</p><h3>Running and Managing the Service</h3><h3>Start/Enable the Service</h3><p>On systemd systems:</p><pre>sudo systemctl start postfwd<br>sudo systemctl enable postfwd</pre><p>Or run manually for testing:</p><pre>sudo postfwd2 -f /etc/postfix/postfwd.cf</pre><p>You can specify user/group with -u and -g if needed.</p><h3>Troubleshooting Common Issues</h3><h3>1. “No rules found — I feel useless”</h3><ul><li><strong>Cause:</strong> Rules file is missing, unreadable, or contains invalid syntax.</li><li><strong>Fix:</strong> Ensure the file exists, is readable, and rules use correct syntax (action=... or action=rate(...) for rate limiting).</li></ul><h3>2. “Rule X contains no action and will be ignored”</h3><ul><li><strong>Cause:</strong> Rule is missing an action=... field, or uses unsupported syntax</li><li><strong>Fix:</strong> Add a valid action, e.g. action=REJECT or action=rate(...).</li></ul><h3>3. Permission Denied on Cache Socket</h3><ul><li><strong>Cause:</strong> User running postfwd2 cannot access the cache socket file.</li><li><strong>Fix:</strong> Start postfwd2 as the correct user, or fix socket permissions. Remove stale socket files if needed.</li></ul><h3>4. Postfix Not Enforcing Rules</h3><ul><li><strong>Cause:</strong> check_policy_service is after permit_mynetworks, so trusted networks bypass policy checks.</li><li><strong>Fix:</strong> Place check_policy_service before permit_mynetworks in main.cf.</li></ul><h3>5. “Unknown smtpd restriction” in logs</h3><ul><li><strong>Cause:</strong> postfwd2 returned an unsupported action string (e.g., using old rcpt=... syntax).</li><li><strong>Fix:</strong> Use action=rate(...) for rate limiting in postfwd.</li></ul><h3>6. Service Fails to Start</h3><ul><li><strong>Check:</strong></li><li>That all Perl dependencies are installed.</li><li>That the config file path is correct (-f /etc/postfix/postfwd.cf).</li><li>That the user/group exists if using -u/-g.</li></ul><h3>Best Practices and Tips</h3><ul><li><strong>Always test your rules</strong> with postfwd2 -C -f ... before deploying.</li><li><strong>Use specific user/group</strong> for running the daemon, not root.</li><li><strong>Keep rules simple and on a single line</strong>.</li><li><strong>Monitor logs</strong> for warnings or errors after changes.</li><li><strong>Document your rules</strong> for future maintenance.</li></ul><h3>Conclusion</h3><p>Postfwd2 is a flexible, powerful tool for enhancing your Postfix mail server’s security and policy enforcement. With the right installation, integration, and rule design, you can implement advanced controls with ease.<br> Always validate and monitor your setup, and consult the <a href="http://postfwd.org/">official documentation</a> for advanced features and updates.</p><h3>Citations:</h3><ol><li><a href="https://github.com/postfwd/postfwd">https://github.com/postfwd/postfwd</a></li><li><a href="https://manpages.ubuntu.com/manpages/focal/man1/postfwd2.1.html">https://manpages.ubuntu.com/manpages/focal/man1/postfwd2.1.html</a></li><li><a href="http://postfwd.org/DEVEL/postfwd2-0.18RC1.html">http://postfwd.org/DEVEL/postfwd2-0.18RC1.html</a></li><li><a href="http://postfwd.org/">http://postfwd.org</a></li><li><a href="https://forum.efa-project.org/viewtopic.php?t=1461">https://forum.efa-project.org/viewtopic.php?t=1461</a></li><li><a href="https://github.com/postfwd/postfwd/issues/15">https://github.com/postfwd/postfwd/issues/15</a></li><li><a href="https://forum.howtoforge.com/threads/postfwd-configuration.86742/">https://forum.howtoforge.com/threads/postfwd-configuration.86742/</a></li><li><a href="https://www.postfix.org/access.5.html">https://www.postfix.org/access.5.html</a></li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=653e4b8d6361" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Installing TrueNAS on Proxmox Using the Web GUI (Q35 + UEFI + Disk Passthrough)]]></title>
            <link>https://medium.com/@sirtcp/installing-truenas-on-proxmox-using-the-web-gui-q35-uefi-disk-passthrough-0f54f09560ba?source=rss-3299d8448ade------2</link>
            <guid isPermaLink="false">https://medium.com/p/0f54f09560ba</guid>
            <category><![CDATA[private-cloud]]></category>
            <category><![CDATA[truenas]]></category>
            <category><![CDATA[gui]]></category>
            <category><![CDATA[proxmox]]></category>
            <dc:creator><![CDATA[Tech Trends Today]]></dc:creator>
            <pubDate>Tue, 06 May 2025 23:41:41 GMT</pubDate>
            <atom:updated>2025-05-06T23:41:41.363Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Overview:</strong> This article walks you through creating a TrueNAS VM in Proxmox via the Web UI. It configures UEFI boot with Q35, adds an EFI disk, and explains how to avoid UEFI boot errors.</p><h3>Step-by-Step Proxmox Web UI Setup</h3><h4>1. Upload the ISO</h4><ul><li>Go to <strong>Datacenter &gt; local &gt; ISO Images</strong></li><li>Upload TrueNAS-23.xx.iso</li></ul><h4>2. Create the VM</h4><ul><li><strong>Create VM</strong> → Name: truenas, VM ID: 110</li><li><strong>OS Tab</strong>:</li><li>ISO Image: select uploaded ISO</li><li>OS Type: Linux (5.x or 2.6 Kernel)</li><li><strong>System Tab</strong>:</li><li>BIOS: OVMF (UEFI)</li><li>Machine: q35</li><li>Check “Add EFI Disk” → Storage: local-lvm</li><li><strong>Hard Disk Tab</strong>:</li><li>Bus: SCSI, Storage: local-lvm, Size: 16GB</li><li><strong>CPU Tab</strong>:</li><li>Cores: 4, Type: host</li><li><strong>Memory Tab</strong>:</li><li>RAM: 8192MB (8GB)</li><li><strong>Network Tab</strong>:</li><li>Model: VirtIO, Bridge: vmbr0</li><li>Click <strong>Finish</strong></li></ul><h3>Post-VM Creation Settings</h3><h4>Set Boot Order</h4><ul><li>VM → Options → Boot Order → Set CD-ROM first</li></ul><h4>Disable Secure Boot</h4><ul><li>VM → Hardware → EFI Disk → Edit → Uncheck Secure Boot</li></ul><h4>Start the VM</h4><ul><li>Go to Console → Start</li></ul><h3>Adding SSDs (Passthrough)</h3><p><strong>Raw Disks:</strong></p><ul><li>Hardware → Add → Hard Disk → Use Physical Disk → Select /dev/disk/by-id/...</li></ul><p><strong>HBA PCI Card:</strong></p><ul><li>Hardware → Add → PCI Device → Select HBA → Enable All Functions if needed</li></ul><h3>Troubleshooting: “Access Denied” on Boot</h3><p><strong>Solution:</strong></p><ul><li>Ensure Secure Boot is <strong>disabled</strong></li><li>Make sure ISO is still mounted</li><li>Set proper boot order</li></ul><p>By following either CLI or GUI method, you’ll have a working, passthrough-ready TrueNAS VM on Proxmox with Q35 and UEFI support.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0f54f09560ba" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>