KrakenHashes Glossary¶
This glossary provides definitions for terms used throughout the KrakenHashes system, organized by category.
Password Cracking Terminology¶
A-Z¶
Association Attack: Hashcat attack mode (-a 9) that tests password candidates against hashes in a 1:1 line-by-line mapping. Hash on line N is tested only against candidate on line N. Useful for targeted attacks with user-specific password intelligence. (v1.4.0+)
Association Wordlist: A wordlist specifically designed for association attacks where each line corresponds to a specific hash in the target hashlist, maintaining line-order correspondence. Line count must exactly match the hashlist hash count. (v1.4.0+)
Attack Mode: The method used by hashcat to attempt password recovery. Common modes include dictionary attack (-a 0), combinator attack (-a 1), brute-force/mask attack (-a 3), hybrid attacks (-a 6, -a 7), and association attack (-a 9).
Benchmark: A test run to measure the cracking speed (hashes per second) of specific hardware against various hash algorithms.
Brute Force Attack: An attack method that systematically tries all possible character combinations within a defined character set and length range.
Candidate: A potential password generated during the cracking process that will be tested against the target hash.
Charset: A defined set of characters used in mask or brute-force attacks (e.g., ?l = lowercase, ?u = uppercase, ?d = digits, ?s = special characters).
Combinator Attack: An attack that combines words from two wordlists to create password candidates (e.g., "password" + "123" = "password123").
Cracked Hash: A hash that has been successfully reversed to reveal its plaintext password.
Dictionary Attack: An attack using a wordlist of common passwords and variations to attempt hash cracking.
Hash: A one-way cryptographic function output that represents a password. Common types include MD5, SHA1, SHA256, bcrypt, and NTLM.
Hash Algorithm: The specific cryptographic function used to create a hash (e.g., MD5, SHA-1, SHA-256, bcrypt, scrypt, Argon2).
Hash Rate: The speed at which password candidates are tested, measured in hashes per second (H/s), kilohashes/s (KH/s), megahashes/s (MH/s), or gigahashes/s (GH/s).
Hashcat: The underlying password recovery tool used by KrakenHashes for distributed cracking operations.
Hashlist: A collection of password hashes to be cracked, typically organized by source, client, or campaign.
Hybrid Attack: An attack combining wordlist entries with masks or rules to generate password candidates.
Increment Mode: A hashcat feature that automatically runs mask attacks for each length from a minimum to maximum value. KrakenHashes decomposes increment mode into discrete layers for distributed processing. See Increment Mode Architecture.
Increment Inverse: Variant of increment mode that processes masks from longest to shortest instead of shortest to longest. Useful when longer passwords are more likely.
Increment Layer: A sub-component of an increment mode job representing one specific mask length. Each layer is scheduled and tracked independently, enabling parallel processing across multiple agents.
Keyspace: The total number of possible password combinations for a given attack configuration.
Mask: A pattern defining the structure of passwords to generate in a mask attack (e.g., ?u?l?l?l?d?d?d?d for Abcd1234 format).
Mask Attack: A targeted brute-force approach using patterns to generate password candidates based on known password structures.
Password Candidate: A potential password being tested against a hash during the cracking process.
Plaintext: The original, unencrypted password that produces a given hash.
Potfile: A file storing previously cracked hashes and their plaintext passwords to avoid redundant work.
Rainbow Table: Pre-computed tables of hash-to-plaintext mappings (not used by hashcat/KrakenHashes).
Rule: A transformation applied to wordlist entries to generate password variants (e.g., appending numbers, capitalizing letters, character substitution).
Rule Splitting: KrakenHashes feature that divides large rule files into chunks for distributed processing across multiple agents.
Salt: Random data added to passwords before hashing to prevent identical passwords from producing identical hashes.
Wordlist: A file containing potential passwords, one per line, used as input for dictionary attacks.
System Architecture Terms¶
A-Z¶
Agent: A compute node running the KrakenHashes agent software that executes hashcat jobs and reports results to the backend.
Agent Pool: A group of agents that can be assigned to work together on jobs.
API Key: Authentication credential used by agents to communicate with the backend server.
Backend: The central KrakenHashes server that manages jobs, stores data, and coordinates agent activities.
Claim Code: A one-time voucher code used to register new agents with the system.
Client: In KrakenHashes context, a customer or engagement for which password cracking services are performed.
Chunk: A portion of work (keyspace segment or rule subset) assigned to an individual agent for processing.
Chunking: The process of dividing large cracking jobs into smaller segments for distributed processing.
Data Retention: Policies and mechanisms for automatically removing old data based on configured time periods.
FIFO Mode: First-In-First-Out agent overflow allocation mode where the oldest job at the same priority receives all overflow agents beyond max_agents limits. Default allocation mode in KrakenHashes.
Heartbeat: Regular status updates sent by agents to the backend to indicate they are alive and processing.
Job: A single password cracking task with specific parameters, wordlists, rules, and target hashes.
Job Execution: An instance of a job being run, which may involve multiple agents and chunks.
Job Template: A reusable job configuration that can be applied to different hashlists.
Job Workflow: A sequence of jobs designed to implement a comprehensive attack strategy.
Max Agents: Maximum number of agents a job can use simultaneously. Respected for jobs at the same priority; overridden when a job has higher priority than all others.
Max Agents Override: Behavior where higher priority jobs receive ALL available agents regardless of their max_agents setting, ensuring critical work completes as fast as possible.
Mixed Work Factors: A condition where hashes in a hashlist have different computational cost parameters (e.g., bcrypt hashes with varying cost values like \(2a\)10$ vs \(2a\)12$). This prevents association attacks since hash order may not be preserved during processing. (v1.4.0+)
Overflow Agents: Agents available beyond the max_agents limits of jobs at the same priority. Distributed according to the agent overflow allocation mode (FIFO or round-robin).
Overflow Allocation Mode: System setting that controls how overflow agents are distributed among jobs at the same priority. Options: FIFO (oldest job gets all) or round-robin (distributed evenly).
Preset: Pre-configured job templates or workflows for common attack scenarios.
Repository Pattern: Software design pattern used in KrakenHashes for database access abstraction.
Round-Robin Mode: Agent overflow allocation mode where overflow agents are distributed evenly among all jobs at the same priority, ensuring balanced progress across multiple jobs.
Service Layer: Business logic layer in the backend that processes requests between handlers and repositories.
WebSocket: Protocol used for real-time bidirectional communication between agents and the backend.
Work Directory: Temporary directory where agents store files and data during job execution.
Work Factor: The computational cost parameter in certain hash algorithms (e.g., bcrypt cost, scrypt N/r/p parameters, Argon2 memory/iterations). Higher work factors increase the time required to compute each hash, making password cracking slower. (v1.4.0+)
Security and Authentication Terms¶
A-Z¶
2FA/MFA: Two-Factor/Multi-Factor Authentication requiring multiple verification methods for user login.
Access Token: Short-lived JWT token used for API authentication.
API Authentication: Token-based authentication system for programmatic access to KrakenHashes.
Backup Codes: One-time use codes for account recovery when primary MFA method is unavailable.
Certificate Authority (CA): Entity that issues digital certificates for TLS/SSL encryption.
CORS: Cross-Origin Resource Sharing - security feature controlling which domains can access the API.
JWT: JSON Web Token - standard for securely transmitting information between parties as a JSON object.
LDAP: Lightweight Directory Access Protocol - external authentication system support.
Rate Limiting: Security measure limiting the number of API requests per time period.
RBAC: Role-Based Access Control - authorization system based on user roles (admin, user, agent, system).
Refresh Token: Long-lived token used to obtain new access tokens without re-authentication.
Self-Signed Certificate: TLS certificate signed by its creator rather than a trusted CA.
Session Management: System for tracking and controlling user login sessions.
TLS/SSL: Transport Layer Security/Secure Sockets Layer - encryption protocols for secure communication.
TOTP: Time-based One-Time Password - MFA method using authenticator apps.
Voucher: Authorization code for specific actions like agent registration or user invitation.
Performance and Optimization Terms¶
A-Z¶
Benchmark Score: Measured performance of hardware against specific hash algorithms.
Cache: Temporary storage of frequently accessed data to improve performance.
Concurrency: Number of simultaneous operations or connections the system can handle.
GPU: Graphics Processing Unit - primary hardware for high-speed password cracking.
GPU Utilization: Percentage of GPU resources being used during cracking operations.
Hash Rate: Speed of password testing, measured in hashes per second (H/s).
Keyspace Distribution: Method of dividing the total keyspace among multiple agents for parallel processing.
Load Balancing: Distribution of work across multiple agents based on their capabilities.
Memory Usage: RAM consumption by hashcat and the agent during operations.
Optimization: Techniques to improve cracking speed or resource efficiency.
Parallel Processing: Simultaneous execution of job chunks across multiple agents.
Performance Metrics: Measurements of system efficiency including hash rate, completion time, and resource usage.
Resource Allocation: Assignment of CPU, GPU, and memory resources to cracking operations.
Thermal Throttling: Automatic reduction in GPU performance to prevent overheating.
Workload Distribution: Strategy for assigning job chunks to agents based on their capabilities.
Job and Task Statuses¶
Job Execution Statuses¶
Pending: Job has been created but has not yet started execution. Waiting for available agents or higher priority jobs to complete.
Running: Job is actively executing with one or more tasks assigned to agents. Agents are processing the job's keyspace.
Paused: Job execution has been temporarily halted by user action. Tasks are stopped but can be resumed later.
Processing: Job has finished execution but is waiting for crack batches to be transmitted from agents and processed by the backend. Progress shows 100% but final completion is pending crack data receipt.
Completed: Job has finished all work, all crack batches have been received and processed, and email notifications (if enabled) have been sent.
Failed: Job encountered an unrecoverable error during execution and cannot continue.
Cancelled: Job was manually cancelled by a user before completion.
Task Statuses¶
Pending: Task has been created but not yet assigned to an agent.
Assigned: Task has been assigned to a specific agent but the agent hasn't started executing it yet.
Reconnect Pending: Task was assigned to an agent that disconnected. Waiting out the reconnection grace period. The task itself is not reassigned — it is closed out (truncated and completed, deleted, or cancelled) and its keyspace range is re-dispatched as a new task. See Task Lifecycle and Statuses.
Running: Task is actively executing on an agent. Hashcat is processing the assigned keyspace.
Processing: Task has completed hashcat execution but is waiting for all crack batches to be transmitted to the backend. Agent has signaled completion but crack data is still being sent. Transient — a background sweep runs every 5 minutes and guarantees a task always leaves this state, terminalising it as Completed or Cancelled (never Failed) if the agent is gone. See Task Lifecycle and Statuses.
Completed: Task has finished execution, all crack batches have been received, and the task is fully complete. Also the outcome of a task that was stopped after making progress: its keyspace interval is truncated at hashcat's restore point and the task is marked completed at 100% of that smaller range, so Completed does not imply the task ran its full original range. See Task Lifecycle and Statuses.
Failed: Usually a failure the agent reported — hashcat could not run, a required wordlist or rule file was missing, and so on. One failed task permanently fails its entire job, even after the re-opened range has been redone successfully, which is why no stop recovery (job stop, preemption, disconnect, heartbeat eviction) ever produces this status. The one server-side exception is retry exhaustion: a task that burns through max_chunk_retry_attempts reconnect or heartbeat retries is failed by the cleanup service with no agent report, and its error_message names the retry exhaustion rather than a hashcat error. See Task Lifecycle and Statuses.
Cancelled: Most commonly a sibling task stopped because the hashlist was fully cracked — which happens on successfully completed jobs, not cancelled ones. Also the result of user action or a job stop, which cancels every running and assigned task outright. Also the benign outcome of a task that was stopped with no usable progress but had already cracked hashes: the row is preserved so crack attribution survives, and its keyspace range is re-dispatched. See Task Lifecycle and Statuses.
Not Listed (task deleted): Not a status. A task stopped with no progress and no cracks has nothing worth preserving, so its row and keyspace interval are both deleted and the task disappears from the job's task list. The whole range re-opens and is re-dispatched. Job task rows are therefore not a complete audit trail of every chunk dispatched. See Task Lifecycle and Statuses.
Common Abbreviations¶
A-Z¶
API: Application Programming Interface
CA: Certificate Authority
CLI: Command Line Interface
CPU: Central Processing Unit
CRUD: Create, Read, Update, Delete (database operations)
CSV: Comma-Separated Values
DB: Database
DNS: Domain Name System
DTO: Data Transfer Object
GPU: Graphics Processing Unit
H/s: Hashes per second
HTTP/HTTPS: Hypertext Transfer Protocol (Secure)
ID: Identifier
IP: Internet Protocol
JSON: JavaScript Object Notation
JWT: JSON Web Token
KH/s: Kilohashes per second (1,000 H/s)
LDAP: Lightweight Directory Access Protocol
MFA: Multi-Factor Authentication
MH/s: Megahashes per second (1,000,000 H/s)
NTLM: NT LAN Manager (Windows password hash format)
ORM: Object-Relational Mapping
OS: Operating System
RAM: Random Access Memory
RBAC: Role-Based Access Control
REST: Representational State Transfer
SHA: Secure Hash Algorithm
SMTP: Simple Mail Transfer Protocol (email transmission protocol)
SQL: Structured Query Language
SSL: Secure Sockets Layer
TLS: Transport Layer Security
TOTP: Time-based One-Time Password
UI/UX: User Interface/User Experience
URI/URL: Uniform Resource Identifier/Locator
UUID: Universally Unique Identifier
VRAM: Video Random Access Memory (GPU memory)
WS: WebSocket
XML: Extensible Markup Language
This glossary is continuously updated as new features and terminology are introduced to the KrakenHashes system.