Chapter 9: Systems administration

Introduction

This introduction frames systems administration work and previews how AI tools such as ChatGPT can simplify recurring administrative tasks throughout the chapter.

• AI-assisted administration use cases: Describe the systems-administration tasks this chapter applies AI to: creating change requests, working with Linux distributions, renaming files, analyzing logs, auditing configurations for compliance, and automating workflows.
  Why it matters: These use cases establish where AI can reduce effort in routine and complex systems-administration work.
• Systems administrator responsibilities: Explain that a systems administrator manages network servers, protects and backs up data, resolves application and service problems, handles user accounts, and may work across Linux and Windows systems or large deployments managed with SCCM.
  Why it matters: Knowing the breadth of the administrator role clarifies why repetitive troubleshooting and configuration work are good candidates for AI assistance.
• Administrative work challenges: Describe why AI assistance is relevant to systems administration: the work includes tedious detailed troubleshooting, complex configurations, and tasks that must be performed repeatedly, while AI can make many of them easier and improve documentation, decisions, and efficiency.
  Why it matters: The value of AI comes from applying it to high-effort, repetitive administrative work rather than using it without a practical purpose.

Key terms: change requests, Linux distributions, log analysis, configuration files, compliance, workflows, systems administrator, servers, Linux, Windows

9.1 Change request

This section shows how ChatGPT can rapidly draft a structured systems-administration change request when given a clear task, while the administrator reviews and adapts the result.

• AI-assisted change request drafting: Describe how to use ChatGPT to draft a formal change request for a systems change, starting with a clear statement of the requested change. Explain that detailed instructions and a well-understood objective help the AI produce a useful first draft.
  Why it matters: A strong AI draft can reduce the time spent on routine change-management documentation.
• Change request structure: Describe the major elements that the AI-generated DHCP migration request includes: the current and proposed states, impact analysis, an implementation plan, a rollback plan, and approval. Explain that these elements document both how the change will be performed and what happens if it fails.
  Why it matters: A structured request makes a proposed infrastructure change easier to evaluate, implement, and reverse safely.
• Human review and adaptation: Explain why the administrator must carefully check AI-generated change-request content, ensure it fits organizational requirements, and adjust it where needed. State that AI provides a useful outline and suggestions, not an unreviewed final authority.
  Why it matters: Reviewing AI output helps ensure that a change plan is appropriate for the organization's actual environment and requirements.

Key terms: change request, ChatGPT, prompt, Current State, Proposed State, Impact Analysis, Implementation Plan, Rollback Plan, Change Advisory Board (CAB), organization’s requirements

9.2 More distros, more problems

This section shows how AI can help administrators account for Linux distribution differences while beginning to diagnose suspected dropped packets on SLES 15.5.

• Linux distribution differences: Explain why administering multiple Linux distributions adds complexity, and identify the kinds of distribution-specific differences that can affect an administrator’s workflow.
  Why it matters: Knowing which details vary by distro helps you avoid applying assumptions from one Linux system to another.
• SLES log investigation: Describe where an administrator can begin looking on SLES 15.5 when slow connectivity and client timeouts suggest dropped packets, including the relevant log locations and the role of local configuration.
  Why it matters: A focused starting point for logs can speed up investigation of a network incident.
• Dropped-packet visibility: Explain how firewall rules and counters can help investigate suspected dropped packets, and why explicit firewall logging may be needed to create monitorable evidence.
  Why it matters: Packet counters and deliberate logging turn a vague network symptom into evidence an administrator can inspect.
• Iterative AI troubleshooting: Describe how the example prompt gives AI enough context to provide a useful troubleshooting starting point, and how an administrator can continue the conversation to investigate the actual firewall configuration or add logging rules.
  Why it matters: Using AI as an iterative troubleshooting partner can help administrators move from symptoms to targeted next steps.

Key terms: Linux distributions, distros, Linux kernel, init system, System V, systemd, SUSE Linux Enterprise Server, SLES 15.5, /var/log, /var/log/messages

9.3 Multiserver administration

This section shows how detailed prompting can generate a Bash script for checking kernel versions across multiple servers and why richer task context produces better AI output.

• Multiserver kernel-check workflow: Describe the AI-generated Bash-script workflow for reading server addresses, connecting to each server without passwords, retrieving its kernel version, and recording a result for every server.
  Why it matters: This replaces repetitive server-by-server checks with a reusable multiserver administration task.
• Kernel version comparison: Explain how the script compares each server's reported kernel version with an expected kernel version and labels the result as lower, equal, or higher.
  Why it matters: A version check is useful only when it turns raw version strings into a clear compliance-style result.
• Detailed role-based prompting: Explain why the detailed request, including the 'Act as a senior Linux system administrator' role and explicit requirements, produced a more sophisticated result than the minimal request.
  Why it matters: Clear context and requirements help a chatbot generate automation that is closer to the administrator's actual needs.

Key terms: Bash script, server list, SSH keys, uname -r, output file, EXPECTED_KERNEL_VERSION, compare_kernel_versions, lower, equal, higher

9.4 Renaming files

This section shows how a plain-English AI prompt can produce a practical PowerShell script for a one-off, conflict-safe batch file and folder renaming task.

• AI-assisted batch renaming: Explain how generative AI can turn a natural-language description of a file-renaming task into PowerShell commands. Describe the chapter-reordering example involving manuscript .adoc files and image folders and files.
  Why it matters: AI can save administrators from manually performing repetitive, highly specific file operations.
• Conflict-safe rename sequencing: Describe why a batch-renaming request must account for name conflicts and may require temporary placeholder names. State that the prompt explicitly asked for this handling when chapter numbers were being reassigned.
  Why it matters: A rename plan can fail or overwrite the wrong item when a desired new name is already in use during the transition.
• Human-resource efficiency: Justify using AI-generated PowerShell for a specific one-time task even when writing a reusable, sophisticated script would take longer than manual work. Explain that the section values a working, efficient solution over unnecessary features such as hashtables and loops.
  Why it matters: For administrative work, optimizing the operator's time can be more valuable than engineering an elaborate script for a task unlikely to recur.

Key terms: PowerShell, generative AI, plain English, manuscript, images, temporary placeholder name, name conflicts, rename files in the subfolders, hashtables, loops

9.5 Analyzing error logs: Identifying exceptions and failures

This section shows how to use ChatGPT to extract critical log messages, diagnose their likely root cause, and protect sensitive data when sharing logs.

• AI-assisted log triage: Explain how uploading a log file to ChatGPT and asking it to extract error, exception, failure, and fatal lines can speed up identification of operational problems and likely root causes.
  Why it matters: AI can reduce the time spent searching large logs so you can focus on resolving the actual problem.
• Sensitive log data protection: Describe the data-protection precautions required before uploading error logs to a chatbot, including using an enterprise provider with data-protection features when possible or removing sensitive identifiers first.
  Why it matters: Logs can expose infrastructure and user information, so protecting them prevents troubleshooting from creating a security risk.
• SharePoint account diagnosis: Trace the SharePoint example from the PostSetupConfiguration TaskException through the validation failure to its likely cause: a local account was used where a domain account was required outside a stand-alone installation.
  Why it matters: Connecting related log entries to one configuration cause helps turn a list of errors into an actionable fix.
• AI-generated remediation guidance: Describe how ChatGPT went beyond locating the critical SharePoint messages by explaining the issue and suggesting remediation: verify the database and account configuration and replace the local account with a suitable alternative when the installation is not stand-alone.
  Why it matters: A useful log-analysis result should help an administrator move from detection and diagnosis toward a concrete next step.

Key terms: log analysis, error logs, root causes, error, exception, failure, enterprise chat provider, data protection features, IP addresses, hostnames

9.6 Automating tasks for efficiency and compliance

This section introduces using AI prompts to incorporate ITIL frameworks and other best practices into systems-engineering automation while maintaining operational efficiency and compliance.

• ITIL integration in automation: Explain why systems engineers need to incorporate ITIL practices into their everyday automation work when their organization adopts ITIL. Describe ITIL as a framework of recommended ways to deliver IT services, rather than as a replacement for automation.
  Why it matters: Integrating required service-management practices into automation helps teams meet organizational standards without abandoning efficient systems work.
• AI prompts for standards: Describe the role of the AI prompts in Table 9.1: they help systems engineers blend frameworks and best practices into their work, simplify complex tasks, and keep automation running while meeting new standards.
  Why it matters: AI can reduce the effort of incorporating changing standards into automation while leaving the automation operational.
• Evolving AI systems role: State the section's forward-looking claim that AI prompts can open the door to AI handling more sophisticated systems-engineering challenges in the future, in addition to simplifying current complex tasks.
  Why it matters: This frames today's prompt-assisted work as a foundation for broader AI support in systems engineering.

Key terms: Information Technology Infrastructure Library, ITIL, ITIL Framework, automation, AI prompts, frameworks, best practices, new standards, systems engineering, Table 9.1

9.7 Auditing configuration files with AI

This section shows how to use ChatGPT to audit configuration files for compliance, security, and performance, using an IIS PCI-DSS review as the main example.

• IIS PCI-DSS configuration audit: Describe how ChatGPT can review an IIS applicationHost.config file against PCI-DSS, identify noncompliant settings, and produce a revised configuration file. State the principal configuration areas the example checks and the changes proposed for them.
  Why it matters: A well-scoped AI audit can quickly surface configuration risks and turn findings into concrete remediation work.
• Actionable audit prompt design: Explain why the IIS audit prompt is effective: it states the compliance objective, asks for noncompliant findings and recommendations, requests a named revised file, uses domain-specific context, and demands actionable output.
  Why it matters: Specific instructions make AI output more useful for a real configuration-review task rather than a vague summary.
• Configuration audit applications: Describe how the same upload-and-prompt approach can audit other configuration-file types for compliance, security, or performance. Give examples from the section and state that the prompt should clearly specify the desired kind of review.
  Why it matters: The approach can be reused across operating systems, web servers, identity systems, containers, and orchestration environments.

Key terms: applicationHost.config, PCI-DSS, SSL/TLS settings, authentication and authorization, logging and monitoring, application isolation, compliance check, recommendations, applicationHost_PCI_compliant.config, naming convention

9.8 Existing AI in systems engineering

This section shows how AI is already integrated into systems engineering tools and highlights current applications for predicting failures, improving disaster recovery, and optimizing network design.

• GPT Actions and API integration: Explain how APIs and GPT Actions let AI models perform specific tasks within common IT software, such as retrieving database information or starting application processes.
  Why it matters: AI integration can move a model from giving advice to carrying out defined actions in operational tools.
• AI systems engineering applications: Describe the three advanced AI applications in systems engineering: predictive system failure analysis, adaptive disaster recovery planning, and network design optimization. Explain the operational benefit of each.
  Why it matters: These applications show that AI can help prevent outages and improve infrastructure decisions, not merely automate routine administration.
• Current AI product examples: State examples of products identified in the section for predictive maintenance, AI-enhanced disaster recovery, and AI-based network optimization.
  Why it matters: Recognizing concrete examples helps connect the use cases to capabilities already available in IT products.

Key terms: APIs, GPT Actions, database information, application processes, predictive system failure analysis, predictive maintenance, advanced disaster recovery planning, enhanced network design, IBM Watson AIOps, Splunk Machine Learning Toolkit

9.9 Prompts used in this chapter

This section collects the prompts used to refine outputs, reformat results, investigate Linux distribution differences, verify table accuracy, and research AI-integrated disaster-recovery products.

• Iterative output refinement: Describe how the prompts refine a successful result by asking how to reproduce it in one shot and by requesting the same output in AsciiDoc format.
  Why it matters: Refining and reformatting a good AI response makes it easier to reuse the result in systems-administration work.
• Distribution variation research: State that one prompt asks what foundational aspects of Linux distributions can vary beyond their use of System V or systemd.
  Why it matters: Considering differences beyond the init system helps avoid treating all distributions as operationally identical.
• Output verification and product research: Describe the prompts that check whether a table introduction matches its contents and ask for commercial disaster-recovery products that integrate AI.
  Why it matters: AI output is more useful when its internal consistency is checked and when research requests clearly specify the required product capabilities.

Key terms: one shot, asciidoc format, distros, System V, systemd, table’s intro, table contents, commercial products, disaster recovery, integrate AI

9.10 Summary

This summary recaps how AI improves systems administration through workflow assistance, scripting, analysis, configuration auditing, and predictive engineering tools.