collective.warmup 1.1

Collective Warmup

Introduction

collective.warmup has been created to warm up web application’s caches upon start and restart.

It works by reading a configuration file containing a list of urls that are then invoked.

collective.warmup is inspired by Gil Forcada’s warmup_plone.py script.

Why not a simple bash script?

While the basic use case is very simple, collective.warmup offers the following extra features:

• Verifies the correctness of the response body (e.g. contains a certain string)

• Can operate in crawl mode, following links in pages to warm up all related pages

• Can filter out certain urls according to a general definition

Installation

collective.warmup can be installed in two ways:

• As a normal python package via pip

• Inside a buildout, with optional integration with Plone

Dependencies

• lxml

Manual installation

To install it (using virtualenv, on a GNU/Debian machine):

$ sudo apt-get install build-essential python-dev python-lxml python-virtualenv libxml2-dev libxslt1-dev
$ virtualenv warmup && cd warmup && souce bin/activate
(warmup) $ pip install collective.warmup

and launch it with:

(warmup) $ bin/warmup <path/file.ini>

Alternative installation with buildout

To integrate collective.warmup in a buildout with Plone, add this to your configuration:

[buildout]
...
parts =
    ...
    warmup


[instance]
...
eggs +=
    collective.warmup


[instance]
environment-vars +=
    WARMUP_BIN ${buildout:directory}/bin/warmup
    WARMUP_INI ${buildout:directory}/warmup.ini


[warmup]
recipe = zc.recipe.egg:scripts
eggs = collective.warmup

After executing the buildout you will find the warmup script in the bin directory.

In this example, the script will be executed automatically by the Zope instance each time it is started.

Configuration file

This is a sample warmup.ini configuration:

[warmup]
enabled = True
sleep = 2
base_url = http://localhost
logfile = /path/to/warmup.log

urls =
    Home page


[config]
max_attempts = 2
base_path = mysite


[Home page]
path = ${config:base_path}/
max_attempts = ${config:max_attempts}
check_exists =
    Welcome
check_not_exists =
    p0wned by
follow_links = True
ignore_middle =
    @@
    ++theme++
    #
    ?
ignore_end =
    .css
    .js
    .png
    .jpg
    .jpeg
    .gif
    .xml
    RSS
    .ico

Options

Global [warmup]

enableTrue or False

If False, the script will do nothing when invoked.

sleepinteger

The number of seconds the script waits between url retrievals. Defaults to 2.

base_urla valid URL (don’t forget the protocol!)

The base URL to check (all paths in the various URL sections are relative to this URL).

log_filea filesystem path

The file where the logs will be written.

urlsa list of sections (separated by newline)

The URLs that we want to check. Each URL must have its own section in the configuration file (see below) and we reference these sections here (do not put real URLs here!). It also set an order for the checks (which are executed sequentially).

URL section

max_attemptsinteger

The maximum number of attempts to check the url. Defaults to 2

paththe path to check

The path will be added to the base_url parameter in order to retrieve the page url

check_existslist of strings

A list of strings that must be present in the page

check_not_existslist of strings

A list of strings that must not be present in the page

follow_linksTrue or False

If True the script will follow the links in the page and will perform the same checks for each link.

ignore_middlelist of strings

If follow_links is True, the links containing one of these strings will be ignored

ignore_endlist of strings

If follow_links is True, the links ending with one of these strings will be ignored

Credits

Changelog

1.1 (2014-07-18)

• Python 2.4 support [gborelli]

1.0 (2014-04-06)

• First release