Architecture / processes¶
This document described the general architecture of IRRd, and some of the most important processes that take place in each component.
As a general process note, IRRd performs all major processes in a single SQL transaction. This includes a single whois query, processing a set of updates from an NRTM server, processing an email with changes to multiple objects, or running a fresh import of a mirrored source. Each of these processes will fail entirely or succeed entirely, which helps to provide consistency in the database.
The basic architecture is this: 1
This is intended as an overview, and not every class is included in this diagram for readability. Significant but out of scope of the diagram are configuration, which is used, for example, by the authentication checker to find the override password, but also by the whois server to determine which port to listen on.
The whois server module deals with whois TCP socket handling and extracting
individual queries. A new thread is started for processing each query.
The parsing is performed by
WhoisQueryParser, which then prepares the
RPSLDatabaseQuery objects. Each TCP connection maps to a single
WhoisQueryParser instance. Most whois queries map to a single
SQL query, but there are a few exceptions. Some state is also kept in
the query parser, such as whether it’s operating in keepalive mode.
IRRd contains a very simple HTTP server for status info. There is only
a single valid URL,
/v1/status. This module contains the HTTP server.
This module receives change requests by e-mail or direct submission. These are first parsed into individual submitted objects, along with their metadata like PGP signatures. The individual object changes are processed by ChangeRequest. This includes retrieving any existing objects, to determine whether the change is an add, update or delete. The new object has its syntax validated, and authentication and referential integrity checks are performed. Note that for one submission, possibly with multiple individual changes, a single authentication and reference validator is used, because their status may on the current validity of other changes in the same message. The ChangeSubmissionHandler resolves these co-dependencies.
The authentication validator will retrieve all relevant mntners for a particular change, and then attempt to pass an authentication method on one of them, given the authentication data provided. In addition, ChangeSubmissionHandler provides it with context on which mntner objects may be new in this same submission, which therefore can’t be found in the database and will require special handling.
The reference validator checks that all referenced objects exist, or in case of deletions, whether there are no more references to the object. Also here, ChangeSubmissionHandler provides context on the current submission, making it possible to delete two objects that depend on each other, in one submission. The information on the references between objects is provided by the RPSL Python objects from irrd.rpsl.
Updates to objects may involve a number of notifications, and a single submission may require notifications to the same or different destinations. For each authentication check, AuthValidator provides all potentially relevant mnt-nfy and upd-to attributes on mntners, ChangeRequest aggregates the correct one of these two options, depending on validation status, along with notify. This is then aggregated in ChangeSubmissionHandler to ensure no more emails are sent than needed.
All objects submitted to IRRd in this way are processed with strict object validation.
The storage module handles any interaction with the SQL database. It offers abstractions to perform complex SQL queries, processing new/updated/deleted objects, and automatic journal keeping. It’s documented in more detail in its own documentation.
The RPSL module handles RPSL objects along with their syntax, metadata, validation and extraction of values like primary keys. It is mainly used when processing imports, NRTM streams and submissions with changes. The particulars of object validation are documented in more detail.
The Python representations of RPSL objects in this module also provide contextual information on the relation between objects. Some, like RPSLMntner, contain additional logic such as validating passwords against a mntner object.
The mirroring module deals with mirroring other databases. This can be done by periodic full imports, or by an initial import followed by an NRTM stream.
A hook to the Twisted framework regularly calls a scheduler, which in turn starts the appropriate mirror update runners. Based on the state of the database and configuration (e.g. is there currently any data? is an NRTM host configured?) the update is either performed by updates from an NRTM stream, or by running a full import. A full import can also be forced by an admin by setting a flag in the database.
Upon each full import, the entire RPSL journal for this mirror is discarded, as the local copy can no longer be guaranteed to be complete.
The configuration module provides other modules access to the IRRd configuration. It loads a default config, followed by the user’s config file. A SIGHUP will reload the configuration, but only after all checks have passed on the new config. However, some settings are only read on startup, and therefore will not take effect until a restart.
The different settings and when they take effect are listed in the configuration documentation.
This module contains scripts intended to be run from the command line.
Most are aimed at development, except
submit_emailwill read an email with updates from stdin and process them. This is the expected entry point for processing incoming email.
submit_changeswill read direct RPSL submissions from stdin and process them. It does not support PGP.
rpsl_readreads files with RPSL data, and inserts them into the database. It is mainly intended for testing, as it does not include aspects like source status metadata updates.
query_qa_comparisonis used for QA tests in query handling.
Note that as separate scripts, they always acts on the current configuration file - not on the configuration that IRRd started with. The latter two scripts are not included in distributions.
The utils module contains a few parts used in other places:
A class for extracting text and PGP metadata from plain and multipart emails, along with other metadata.
A wrapper around gpg to validate an inline or PGP/MIME signature.
Sample RPSL objects used for tests.
Text utilities for working with RPSL paragraphs and lines. Notably, when separating RPSL object lines, Unicode newline characters must not be considered newlines, which is contrary to the behaviour of built-in Python functions.
A small whois client, used by the status info page.