Linux Server Admin with Fexingo: Sysadmin, Bash, and Server Engineering

In this episode of Linux Server Admin with Fexingo, Lucas and Luna dive into AIDE (Advanced Intrusion Detection Environment) for monitoring config file drift on Linux servers. They walk through a real example: a misconfigured Apache virtual host that went unnoticed for weeks after a junior admin changed a DocumentRoot path without updating the change log. Lucas explains how AIDE builds a baseline database of file hashes, then runs periodic checks to flag unauthorized modifications. They cover initialization, configuration /etc/aide/aide.conf, rule groups like 'Normal' and 'DataOnly', cron scheduling, and what to do when AIDE alerts on an expected change. The episode also touches on AIDE's database storage options and the tension between security and operational noise. Perfect for sysadmins who want to catch configuration drift before it becomes an outage.

#AIDE #AdvancedIntrusionDetectionEnvironment #ConfigDrift #LinuxServer #Sysadmin #Bash #ServerEngineering #FileIntegrityMonitoring #Apache #Cron #DevOps #Security #ITInfrastructure #LinuxSecurity #FexingoBusiness #BusinessPodcast #Technology #LinuxAdmin

Keep every episode free: buymeacoffee.com/fexingo

In this episode of Linux Server Admin with Fexingo, Lucas and Luna dive into Linux server I/O scheduling — a critical but often overlooked performance lever. They break down the three main schedulers in the modern kernel — mq-deadline, kyber, and BFQ — and explain when to use each one based on your workload. Lucas shares a real-world example: tuning a database server's NVMe drives from the default kyber to mq-deadline, which cut read latency spikes by 30 percent. Luna asks about the old CFQ scheduler and why it was removed. They also cover how to check the current scheduler per block device, change it at runtime with a simple echo command, and make the change persistent via udev rules. The episode ends with a brief appeal for listener support to keep the show ad-free, tied to the value of practical, no-fluff sysadmin knowledge.

#Linux #Sysadmin #ServerPerformance #IOScheduling #Kernel #BlockLayer #mqDeadline #Kyber #BFQ #StorageTuning #NVMe #Udev #PerformanceTuning #Technology #FexingoBusiness #BusinessPodcast #ServerAdmin #LinuxAdmin

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna dive into automated security patching for Linux servers using the unattended-upgrades package. They discuss why many sysadmins fear automatic updates, how to configure unattended-upgrades safely for production environments, and the real-world experience of a company that reduced their patch window from two weeks to under four hours without a single outage. Lucas explains the key configuration options, including pinning packages, setting up email notifications, and integrating with monitoring tools like Nagios. Luna challenges the approach with concerns about testing and regression, and they explore strategies like canary servers and staged rollouts. By the end, you'll have a concrete plan to implement automated patching while maintaining control and observability.

#LinuxServerAdmin #Sysadmin #UnattendedUpgrades #AutomatedPatching #SecurityUpdates #APT #Debian #Ubuntu #PatchManagement #ProductionDeployments #CanaryDeployments #RegressionTesting #Nagios #DevOps #ServerEngineering #Bash #Technology #FexingoBusiness

Keep every episode free: buymeacoffee.com/fexingo

In this episode of Linux Server Admin, Lucas and Luna dive into the practical art of network troubleshooting using two often-overlooked command-line tools: tracepath and MTR (My TraceRoute). They explain how tracepath provides a quick, no-frills path discovery without requiring root privileges, while MTR combines the best of traceroute and ping in real time, giving system administrators a powerful view into packet loss and latency per hop. The conversation focuses on concrete scenarios—diagnosing a slow connection to a remote server, identifying a misconfigured firewall dropping ICMP packets, and distinguishing between ISP routing issues and server-side problems. Lucas shares a real-world story from a Colo data center outage where MTR pinpointed a failing switch port that traceroute alone missed. Luna asks sharp questions about rate limiting, false positives, and when to reach for tcpdump instead. By the end, listeners will walk away with a clear, repeatable workflow: start with tracepath for a quick sanity check, escalate to MTR for deeper analysis, and correlate results with system logs. Perfect for sysadmins who manage remote servers and need to diagnose network issues fast without relying on SaaS monitoring tools.

#Linux #Sysadmin #NetworkTroubleshooting #Tracepath #MTR #MyTraceRoute #ServerEngineering #CommandLine #ICMP #PacketLoss #Latency #Routing #DevOps #Technology #FexingoBusiness #BusinessPodcast #LinuxServerAdmin #Bash

Keep every episode free: buymeacoffee.com/fexingo

In episode 115, Lucas and Luna break down the practical decisions behind RAID configuration on Linux servers. They walk through the trade-offs between hardware and software RAID, explain when mdadm still beats modern alternatives, and share a real-world story about a misconfigured RAID 5 array that cost a startup two days of downtime. The episode covers stripe size selection, the case for RAID 10 over RAID 5 in database workloads, and how to monitor array health with smartmontools and mdstat. If you manage Linux storage, this episode gives you a clear framework for choosing the right RAID level and avoiding common pitfalls.

#LinuxServerAdmin #RAID #Mdadm #StorageManagement #LinuxRAID #RAID10 #RAID5 #StripeSize #Smartmontools #Mdstat #DataRedundancy #ServerEngineering #Sysadmin #Bash #Technology #FexingoBusiness #BusinessPodcast #ServerStorage

Keep every episode free: buymeacoffee.com/fexingo

In this episode of Linux Server Admin with Fexingo, Lucas and Luna dive into securing NFSv4 with Kerberos authentication. They walk through the practical steps of configuring a Kerberos Key Distribution Center (KDC) on CentOS Stream, setting up NFSv4 with RPCSEC_GSS, and debugging common issues like keytab mismatches and time sync problems. By the end, you'll understand how to replace insecure AUTH_SYS with strong Kerberos-based security for your network file shares. Perfect for sysadmins tired of relying on IP-based access control for NFS.

#LinuxServerAdmin #NFSv4 #Kerberos #RPCSECGSS #SystemAdministration #CentOSStream #KDC #NetworkSecurity #Keytab #GSSAPI #Ticketing #Chrony #FexingoBusiness #BusinessPodcast #Technology #Sysadmin #ServerEngineering #Bash

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna dive into Linux server storage tiering using LVM cache, a technique that combines fast SSDs with slower HDDs to create a hybrid volume. They explain how LVM cache works on a block level, using the 'cache' target in device-mapper to promote frequently accessed blocks to the fast tier. Lucas walks through a concrete example: setting up a cache pool with an NVMe drive backing a 4 TB HDD, using practical commands like 'lvcreate --type cache-pool' and 'lvconvert --type cache'. They discuss the two cache modes—writethrough for safety and writeback for performance—and when to choose each. Luna probes the tradeoffs: metadata overhead, write amplification on the SSD, and failure scenarios like a cache device dying. They also touch on newer alternatives like bcache and Stratis, and why LVM cache remains a solid enterprise choice for mixed-media storage. If your servers have both fast and slow drives, this episode gives you a clear path to better utilization without additional software.

#Linux #ServerAdmin #LVM #StorageTiering #Cache #DeviceMapper #SSD #HDD #Sysadmin #Bash #ServerEngineering #Technology #FexingoBusiness #BusinessPodcast #LinuxServer #StorageOptimization #Lvmcache #EnterpriseStorage

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna dive into Intel QuickAssist Technology (QAT) for Linux servers—a hardware accelerator that offloads cryptographic operations like TLS handshakes and IPsec encryption from the CPU. They walk through how QAT works at the kernel driver level, how it integrates with OpenSSL via the async engine, and real-world performance gains: up to 50% lower latency for TLS and 4x throughput for symmetric encryption on the same hardware. The hosts discuss when QAT makes sense (high-traffic web servers, VPN gateways, CDN nodes) and its trade-offs: increased power consumption, licensing costs, and the need to pin application memory for DMA. They also touch on alternatives like AES-NI and ARM's Cryptographic Extension. This is a practical guide for sysadmins considering hardware crypto offload to scale without upgrading CPU cores.

#IntelQAT #QuickAssistTechnology #LinuxServer #Cryptography #HardwareAcceleration #TLS #IPsec #OpenSSL #Sysadmin #PerformanceTuning #ServerOptimization #KernelDrivers #DMABuffer #AESNI #Tech #Linux #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

In this episode of Linux Server Admin with Fexingo, Lucas and Luna dive into setting up a Samba-based Active Directory Domain Controller on a Linux server. They cover the key differences between Samba 4 and classic NT4-style domains, walk through the necessary packages and configuration steps for joining a domain, and discuss real-world use cases like centralized authentication and group policy management. The hosts also highlight common pitfalls such as DNS integration and time synchronization. Whether you're managing a mixed Windows-Linux environment or consolidating identity services, this episode provides a practical anchor for deploying Samba AD DC in production.

#LinuxServerAdmin #Samba #ActiveDirectory #DomainController #IdentityManagement #WindowsIntegration #LDAP #Kerberos #DNS #NetworkAuthentication #Sysadmin #Bash #ServerEngineering #Technology #FexingoBusiness #BusinessPodcast #OpenSource #Infrastructure

Keep every episode free: buymeacoffee.com/fexingo

Ever had a production application fail silently with 'permission denied' and spent hours chasing ghosts? Lucas and Luna break down strace, the Linux system call tracer that saved one sysadmin's weekend after a custom daemon crashed on startup. They walk through a real example: tracing execve, open, and connect syscalls to spot missing file descriptors and misconfigured seccomp filters. You'll learn the three flags Lucas reaches for first (-f for follow forks, -e for event filtering, -o for output files) and why strace can be your fastest path from 'it doesn't work' to 'ah, it's in the audit log.' No fluff, just the tracer that sees everything.

#Linux #Sysadmin #strace #SystemCallDebugging #ServerTroubleshooting #DevOps #Debugging #LinuxKernel #Syscalls #Seccomp #Auditd #ServerAdmin #Technology #FexingoBusiness #BusinessPodcast #Bash #ServerEngineering #LinuxServer

Keep every episode free: buymeacoffee.com/fexingo