Ch6: Models of network and system administration
Introduction to Network and System Administration
System Interaction:
Systems consist of interacting components that have mutual dependencies.
Ignoring these dependencies results in management errors and false assumptions.
Systems operate as complex cause-and-effect relationships, comparable to ecological webs.
Adaptability:
An adaptable system is defined by its ability to cope with unexpected changes.
If original assumptions about a system fail, they can be modified in an adaptable system.
Adaptability enhances predictability during changes and facilitates recovery from failures.
The Role of System Management:
The primary role of management is to secure the conditions required for system components to perform their specific functions.
Management should avoid controlling every minor detail; the focus must remain on securing necessary conditions.
Systems are first designed to function, and then management discusses attributes based on three key themes:
Resource Management: Managing consumables and reusables.
Scheduling: Managing time and queues.
Strategy.
6.1 Information Models and Directory Services
Definition and Function:
A directory service is a software application designed to manage information regarding network resources.
It acts as a centralized database to organize and store data about computers, users, printers, and various network devices.
Access is restricted to authorized applications and users.
Utility in Enterprise Networks:
Widely utilized for authentication, authorization, and general resource management.
Enables Single Sign-On (SSO) capabilities, allowing users to log in once and access resources across multiple systems without repeated authentication.
Named Examples:
Active Directory (Microsoft)
OpenLDAP (The OpenLDAP Project)
Novell eDirectory (Micro Focus)
Usage Models:
White Pages: A lookup performed when the user starts with a specific key for a specific resource (e.g., looking up a specific number in a telephone book).
Yellow Pages: A browsing-based service used when the exact resource is unknown, requiring a list of categories to match services or users.
Technical Infrastructure:
Maps network resources to their respective physical network addresses.
Provides a shared infrastructure for organizing and administering resources such as volumes, files, folders, groups, users, devices, and telephone numbers.
In a directory server (or name server), every network resource is treated as an object.
Information about resources is stored as a collection of attributes associated with that object.
Namespace and Identification:
Directory services define a network namespace and assign a unique identifier (name) to every object.
Namespace rules ensure names are unique and unambiguous.
Allows users to locate resources by name instead of physical address.
Access control provisions may limit information availability to authorized users.
X.500 Standards and Protocols
Overview:
X.500 refers to a set of ITU-T standards defining protocols for global directory services.
It uses a hierarchical directory structure for organizing network resource information.
While widely used in enterprises for auth/auth and resource management, it has largely been superseded by the simpler Lightweight Directory Access Protocol (LDAP).
Protocol Suite:
Directory Access Protocol (DAP)
Lightweight Directory Access Protocol (LDAP) (Note: DAP and LDAP constitute Group 1; they differ in description and complexity)
Directory System Protocol (DSP)
Directory Information Shadowing Protocol (DISP): Used as a replication protocol.
Directory Operational Bindings Management Protocol (DOP)
Certificate Authority Subscription Protocol (CASP)
Authorization Validation Management Protocol (AVMP)
Trust Broker Protocol (TBP)
Object Classes:
Abstract: Located at the upper levels of the hierarchy. Entries are only populated if inherited by a structural class. Examples include ‘top’, ‘Country’, ‘Device’, ‘Organizational-Person’, and ‘Security-Object’.
Structural: The primary content or ‘meat’ of an object class used for actual entries. Examples include ‘person’ and ‘organization’. The attribute ‘objectClass’ declares the entry type (e.g., ‘Computer’ or ‘Configuration’).
Auxiliary: Defines special-case attributes added to specific entries as hints. For instance, both a person and an organization might need a telephone number or web page, even if it is not a requirement of their structural class.
Distributed Directory Partitioning:
As shown in Figure 6.1, a distributed directory can be partitioned into distinct areas, each handled by a separate server (indicated by dotted areas in the visualization).
6.2 System Infrastructure Organization
Economic Balance: Administrators must find the equilibrium between spending money effectively and saving money where appropriate.
Team Work and Communication:
Bi-directional communication is essential in several forms:
Computer programs and their data.
Computers and devices.
Collaborating humans in teams.
Clients and servers.
Users and systems.
Policy decision-makers and policy enforcers.
Computers and the environment (indicated by physical hazards like spilled coffee).
Communication Noise and Errors:
Environmental noise can interfere with communication in two ways:
Faulty Communication/External Interference: Information is distorted, omitted, or inserted.
Incorrect Interpretation: Symbols are incorrectly identified due to external interference or imprecision.
System Administration as a Meta-Program:
System administration is viewed as a meta-program executed by both machines and humans to manage system evolution. It involves:
Configuring systems within policy.
Maintaining machine operation within policy.
Managing user activity within policy.
The Development Loop (Figure 6.2):
Visualizes system development over time based on a set of rules.
Components: Rule, Message, Noise, Computer, and Users.
Users influence the system by altering rules, changing the conditions for rules, or directly touching/configuring the computer.
Homogeneity and Delegation
6.2.2 Homogeneity (Uniformity):
Refers to making all hosts identical or similar in software and hardware configuration.
Advantages: Increased system predictability, easier management for administrators, and the ability to reuse hardware during emergencies.
Disadvantages: Specialized hardware may offer superior performance for specific applications (e.g., high-availability servers requiring multiple processors, high memory, and high bandwidth) that homogeneous systems lack.
Delegation II:
Assigns responsibilities to local administrators who understand local host usage patterns.
Minimizes the distance between the administrative center and the zone of responsibility.
Types of Delegation:
Administrative: Assigning specific tasks to a subordinate.
Project: Responsibility for a specific project/set of tasks.
Resource: Access to budget or equipment.
Authority: Delegating the power to make decisions on behalf of a superior.
Policy: Responsibility for creating, implementing, and enforcing policies.
Functional: Delegating specific functions (e.g., HR or marketing).
6.2.4 Mobile and Ad Hoc Networks
Definition: An Ad Hoc Network (AHN) is a collection of mobile objects capable of transmitting information.
Characteristics:
Forms an arbitrary graph that evolves over time.
Nodes (humans and devices) move randomly, leading to rapid and unpredictable changes in network topology.
Important for intermittent connectivity and describing high-level associations (who is in contact with whom; information flow paths).
Study Importance:
Flexibility: Adapts to changing topologies; ideal for military operations, emergency response, or remote locations.
Resource Efficiency: No need for pre-existing infrastructure; cost-effective and quick to deploy.
Scalability: Can scale from small personal networks to large industrial ones.
Research & Development: A fresh field leading to new technological advancements.
6.2.5 Peer-to-Peer (P2P) Services
Definition: Applications where each node chooses whether to participate in or abstain from data exchange with others over a common channel.
Key Concepts:
Platforms connect parties directly without a third-party intermediary.
Technology is used to overcome traditional transaction costs related to trust, enforcement, and information asymmetries.
Services provided include payment processing, quality assurance, and information about buyers/sellers.
Examples:
Open-source Software.
Filesharing.
Online Marketplaces.
Cryptocurrency and Blockchain.