Survol server-side code is a pure-Python module and as such can be installed like any module. The packages are available on the Python Package Index (Pypi), which is a repository of software for the Python programming language.
The installation command is:
It is also possible to install Survol from the source code which is available at two places:
Your machine needs a recent Python 2 or Python 3 interpreter. Great care is taken to ensure compatibility with all common Python versions.
The required Python modules are:
There are no other requirements. The rest is optional.
Survol can of course be installed from the sources, which are stored at two places:
Survol comes with a very simple CGI server started by the script survol/scripts/cgiserver.py . Therefore, it can run on very small platforms. Also, it is easy to run Survol on a user account with full read access rights.
$ cgiserver - h
Survol HTTP server:
cgiserver -
p, --port = < number > TCP / IP port number.Default is 8000 -
a, --address = < host > IP address.Default value is local host name. -
b, --browser = < program > Starts a browser -
v, --verbose Verbose mode
Example of commands:
$ cgiserver - b "C:\Program Files (x86)\Mozilla Firefox\firefox.exe"
$ cgiserver - b "C:\Program Files\Internet Explorer\iexplore.exe"
$ cgiserver - b webbrowser# Uses default browser. If your browser runs ActiveX, it is also possible to use Survol web pages with no agent.
Survol is based on a tree of Python scripts, all of them displaying information of the system the agent is running on. Some scripts do not need parameters: they will return information with no context: for example, all available databases, all detected machines on the network, all installed Python modules etc... Other scripts need parameters, as they display information about a specific object. For example; Files opened by a specific process. Columns of a SQL Server™ table. Processes connected to an Oracle™ database.
The HTTP Server Apache must be able to run all Python scripts residing under the directory survol/sources_ types as CGI scripts. On Windows, it is necessary to specify to Apache which Python interpreter must be used to run the CGI scripts. On Windows, this is done in the registry, with various keys depending on the Windows operating system version.
On Windows 7, see the key HKEY_CLASSES_ROOT\.py\Shell\ExecCGI\Command
: The key "Default" contains the program name, which can
for example be:
C:\Program Files (x86)\Python32\pythonw.exe -u
With Windows XP, the command is the key itself.
"C:\Python\3.2.3-0.3\pythonw.exe -u"\
On Linux, all CGI scripts must be executable
and the first line specifies the Python interpreter.
The configuration file httpd.conf
must be edited. It is also possible to create a new survol.conf
file, which is a preferred solution. In both cases, the content is
similar to these lines, which suppose that the sources
are copied in the
directory.
C:/Projects/Survol:
AddHandler cgi-script .py # Depending on where the sources are:
Alias /Survol "C:/Projects/Survol"
<Directory "C:/Projects/Survol">
Options Indexes FollowSymLinks Includes ExecCGI
Allow from all
AddHandler cgi-script .py
ScriptInterpreterSource Registry-Strict
SetEnv PYTHONPATH C:/Projects/Survol/survol
</Directory>
All HTTP results generated by scripts allow HTTP
cross requests (Access-Control-Allow-Origin).
This permits to merge information from several agents, into a
single browser session. It makes also possible for a
static html page to load json URLs from remote agents.
Once the core features are installed, it might be possible that some CGI scripts do not work, or not as expected. Errors are normally displayed when detected, and the Apache log file contains all the information written to stderr. If this is not enough, a very powerful debugging technique is to run CGI scripts as command- line Python mode, by setting a couple of mandatory environment variables. This allows to test a program in isolation, possibly in a debugger, with full visibility on its context, inputs and outputs. Here is an example:
PYTHONPATH=/homez.85/primhilltc/survol/survol \
SERVER_NAME=debug.primhillcomputers.com \
SCRIPT_NAME=sources_types/CIM_ComputerSystem/mysql_databases.py \
QUERY_STRING="xid=CIM_ComputerSystem.Name%3Dprimhilltcsrvdb1.mysql.db&mode=json" \
python survol/sources_types/CIM_ComputerSystem/mysql_databases.py
With this command, the CGI script mysql_databases.py
runs as if it is launched by a HTTP server.
Service Location Protocol (SLP) defined by the Internet Engineering Task Force (IETF) allows networking applications to discover the existence, location, and configuration of networked services. It is already used by WBEM, and also by Survol, for the same reasons. The chosen implementation for Survol, is OpenSLP, which has been ported to a wide variety of systems. It is run with the slptool command from sub-processes. This prevents linking issues, 32bits/64bits compatibility problems etc... However, slptool command must be accessible from the Python scripts.
It might be helpful to pre-register some services in the file slp.reg. This is the right place to also register manually, WBEM services. An example of configuration is:
#
service:wbem:http://mymachine-hp,en,65535
description=OpenPegasus under Windows 7
# Definitions must be separated by an empty line.
service:survol:http://mymachine-hp:8000/survol/entity.py,en,65535
description=Survol Windows 7
#
SLP needs multicast, therefore some specific tuning network parameters and routers might be necessary. Because SLP is a standard, Survol is able to discover all WBEM service exposed with any implementation of SLP.
