Use this directive to disallow overriding of configuration settings made in the main configuration files for just the given directive, regardless of what the default behaviour given in AccessPolicy is.
Use this directive to allow settings from the main configuration file to get merged with the ones found in .nwaccess files for the directive named here, regardless of the default behaviour given by AccessPolicy.
This enables overriding of settings made in the main configuration files with the ones found in .nwaccess files for the directive name given here, regardless of the default behaviour given with AccessPolicy.
Defines the default access files policy regarding overriding of configuration
values from higher level configuration files (for example .nwaccess file in
another directory). You can override the default behaviour defined here
for some directives using AccessMerge,
block means that, configuration values cannot be overriden by settings in a .nwaccess file. Setting override here allows that per default. merge defines that settings from the main configuration files are merged with the ones found in .nwaccess files.
Gives the prefered RBL query method. The default "dns" is much faster than
the http access method "web".
See also mail-abuse.org/rbl for further informations on the RBL system.
Note: The maps service is free for personal usage only; but you can easily edit this nanoweb modules` code to use another one if desired.
This directive sets the default access control policy. For public webservers
you may wish to leave this value to allow, for intranet systems
however deny is recommended.
The default behaviour can be tweaked for individual hosts using ACAllowIP, ACAllowHost, ACDenyIP and ACDenyHost.
These directives effectively allow the specified hosts to access files from your webserver. You may give complete IP addresses and host names for a single host here as well.
Defines the response code for hosts which you denied access to the server.
You can also set an individual greeting for hosts that have no permission to access the server.
Using this directives you can deny the specified hosts access to files from your webserver, even if the default access control policy was set to 'allow'.
You can append additional headers to the HTTP responses nanoweb sends back
to the client (browser). This is a very powerful feature if you know the
headers recognized by special http clients (or you've written one of your
Using the AddServerVar directive you can pass arbitrary environmental
variables to any executed CGI program. You may also overwrite one of the
default server vars with any static value you
would like to (so you can fault any CGI, for example let it assume a
different webserver or tell it a false request uri).
Don't use an equal sign between the variable name (uppercase letters are recommended) and its new value! You may use this directive in .nwaccess files, too, of course.
This directive allows you to add a mime type for an extension on the fly without modifying your systems` mime.types file, it can however be used in the main server configuration file only.
Declares an alias directory name to an existing path of the real
filesystem. This directive can be used to get access to a directory
from within any virtual host, as in the rewriting process the docroot
This directive is used by mod_fb for example, to always keep the /icons/ directory reachable.
The newer AllowSymlinkTo directive allows now much finer grained configuration of which symlinks are allowes within the docroot.
Nanoweb will generelly reject symlinks to outside of the
DocumentRoot for security reasons.
Using this directive you can give destination paths where symlinks are
however allowed to (the directive can be given multiple times).
If you just set it to the root directory / symlinks will always be allowed.
Any CGI may be called with a »pathinfo« appended to its filename (think of http://www.example.com/cgi-bin/program.php/arg1/arg2), which is treaten by many CGI interpreters as additional data input beside the »query string«.
You must specify this directive before the LoadModule directive for mod_auth_anydb, if you want to use the ADOdb or the PEAR::DB database abstraction layer.
With the ANYDB authentication method you can allow access to a restricted
area using the auth data (login/user names and passwords) from any PEAR::DB,
dbx or ADOdb supported database.
The syntax for this directive is often called "dsn" (data source name), but additonally contains the name of the database table with the login names and passwords. The dbtype part tells which RDBMS you use, but the value depends highly on the database abstraction layer you choose (PEAR, ADO, dbx).
The ANONYMOUS authentication method allows to login suppling a email address as password for either "anonymous" or any other symbolic user name specified with this directive. You can give additonal names in one line or within multiple directive lines.
The email address given as password for ANONYMOUS http authentication logins can be verified, and if the quick SMTP check fails, the mail address is rejected and cannot be used to access the restricted area.
This directive was used until 1.9.0, and is called AuthNwauthFilename now and used by mod_auth_nwauth.
With the HTPASSWD authentication method you can reuse the password files from the apache webserver, these are often called .htpasswd
Using the AuthLocation in the global, virtualhost, or access file will restrict the need for authentication to the specified locations.
You can give just ".nwauth" to this directive to activate the authentication with "AuthRequire=NWAUTH" from a directories .nwaccess
These directives specify the mysql database + server, the database table and the columns which hold the user names and passwords to check for when using the MYSQL authentication method. The PassType setting tells if the password is saved unencoded in the table (plain) or using one of the system password hash algortithms (crypt, md5) or the according MySQL function (mysql).
Use these directives to give the name of the database of the PostgreSQL server with the column names which hold the user names and passwords to check for when using the PGSQL authentication method shall be used (see AuthRequire). The PassType setting tells if the password is saved as plain text in the table or was encoded using md5()
The SIMPLE authentication method allows you to specify the login and
password pairs directly in the config file (.nwaccess most often, but may be
also in one of the vhost sections to restrict access to the whole server).
You need to give one directive with space seperated login name and password for each user, most often in the same config file where you activated the authentication (with AuthRequire).
This directive is to be used to give the different authentication areas
(most often single directories or directory trees) a name, to help users to
remember the correct login and password to be granted access.
Most often this is shown as info text in common browsers, so you may want to give a notice or a greeting message here.
The different modules give you different choices on where to keep the
authentication passwords and login names, so you need to specify which auth
module to use for the current directory (or server if you want to restrict
one of your vhosts at whole).
Currently there a 6 different auth modules (have a look into the modules section), and thus you can specify SIMPLE, HTPASSWD, NWAUTH or the MYSQL, PGSQL, LDAP authentication method/module here.
The AuthRequire is the directive out of the bunch of authentication directives that really triggers if the access to a directory is restricted, so you may specify all your passwords (say with AuthSimpleUserPass directives) in the main server configuration file but activate it occassionally in the .nwaccess files of some directories only.
You can enable more than one authentication method by using this directive two or more times in one of your .nwaccess files. This leads to a logical OR, which means that the users login name and password needs to be verified by one of the authentication modules only to grant him or her access.
Limits the average download throughput to the given fixed amount for the
server area it was specified ([global], [vhost] or a single directory via
As the module only estimates the transfer size for every second, so the "/s" is optional (because the time slices cannot be influenced). You can however give the speed amount in bytes or in kilobytes (with an appended "K").
Allows to view the source of .bsp scripts when ?source is appended to the requested url.
If you plan to use the CGI version of PHP you may want to leave this directive enabled, as there still is a unfixed bug in PHP-CGI that forbids passing the $PATH_INFO server var to the scripts as this would trash the $PHP_SELF var. If you don't use the latter one in your scripts you otherwise should probably disable this option, and use the more reliable $SCRIPT_NAME in favour of $PHP_SELF.
This directive tells nanoweb, that the specified directory contains CGI
scripts only. It is very common to have at least one such directory, which
is often called /cgi-bin/ (and accessible with this name by using the
If you do not want to have just one seperate directory for all your CGIs, you may set this directive to / to allow files located anywhere on your server to be treated as CGIs if they have the executable flag set.
If a file in a CGIScriptsDir is accessed which is not marked executable, this directive tells nanoweb how to treat this. The value »error« leads to an error response message if a script without exec-bit is found, »raw« will deliver the according file as is (this is discouraged because of security reasons).
You may override php.ini directives on a per directory basis with this
directive. Don't put any spaces in between the equal sign and the option
name and its value. If the value contains spaces you have to enclose it in
quotation marks. For settings that expect on/off in php.ini please use 0 and
This directive corresponds to apaches php_flag, php_option, ... directives.
The absolute maximum time (in seconds) a child process is allowed to live.
Enabling this will add the HTTP header Content-MD5: for static
responses, which contains the md5() hash summary of the served file to
allow the client to verify correct transmission.
This requires the mod_digest extension module.
When the server cannot determine a MIME-type by examining the extensions of the requested file, it will tell the client the file is of the type set with this directive. The default of text/plain may not always suit your requirements, so setting this value to text/html can do a better job in some cases.
Nanoweb feeds all files internally through an "parser"; you may specify one of them here (say CGI or SSI) if desired, but it is very rarely a good idea to feed standard static files through the PHP parser for example.
With this directive you may specify one or more files, which get sent in
favour of a directory listing when found - this is very often called the
»start file« and index.html per default. You can alternatively
define the absolute path to a cgi script, which will then be activated to
generate a directory listing (instead of Nanowebs
mod_fb) if none of the other index files exists.
The value of this directive may also be overriden finally in the per-directory configuration files (.nwaccess).
You may want to set it to just "index" if you activated mod_multiviews, as this is more flexible.
If nanoweb cannot determine the mime type from a files` extension then the
PHP4.3 function mime_content_type() will be engaged to analyze the first 100
bytes of a file to guess the correct mime type. This directive can be used
to disable this behaviour if you fear it slows down your server too much.
If disabled Nanoweb will always report the DefaultContentType if a extension is unknown. You must set the according php.ini option if your mime-magic.data file is not located at its usual place.
Base directory path nanoweb will serve files from and where you should put
all your html files and images into, as nanoweb doesn't allow to deliver
files which are not located in there.
This directive needs to be present in each virtual host configuration section.
Sets the http response code for »Denial of Service« attacks.
The maximal number of requests one host may make in the time slice specified with DosEvasiveTimer before it is considered to be an attack.
Gives a delay in seconds that is used to detect denial of service attacks.
Defines files which get send on errors instead of the default (built-in) nanoweb messages. These documents must reside in the docroot.
You can attach additional HTTP headers to responses with an error code above 400; see also AddHeader.
If the mime qualities of all available file variants are equal, then mod_multiviews can at least select the smallest, largest, newest or the oldest file out of them.
Descriptions of each file in a directory may be given in the file
specified here, its content is used to enhance the output of directory
listings produced by the file browser module.
Such a description file should look like that:
filename.ext desc of this file index.html main/homepage...
Associates icons pictures to mime types for output in directory listings
produced by mod_fb.
You may specify partial mime-types only; so you can associtate fall-back icons for general mime types like »application«, »text« or »image«.
The default configuration of nanoweb also utilizes the Alias directive to keep all icons in a single directory (/icons/ which is rewritten to absolute real path /var/www/icons/) accessible from all the virtual hosts.
If none of the FBIconByType directives defines an appropriate icon for a file (or if a files mime type is unknown) this directive gives the name of the image to be used.
Defines the icon to be printed for directories.
This tells mod_fb to include files which names start with a dot to be included in directory listings. On UNIX machines these »dotfiles« should be treaten as invisible files by standard programs; and because nanoweb keeps some of its per-directory configurations in .nwaccess and .nwauth you probably want to set this directive to the default of 1.
Directory listings are sorted by mod_fb according to this setting (»name« is the default). The use of »desc« reverses the order.
Dropped in favour of ServerThemes.
Removed in favour of ServerThemes.
Whenever a directory listing (»filebrowser«) is to be sent back to the client, the server searches for a file with the hereby given name to be included in the sent output.
See CGIFilterPathInfo on informations about this directive. The name of this directive was renamed from FastCGIFilterPathInfo in 2.1.0 to normalize its name with all the other ones.
Strips the given header from the output of a CGI.
Is to activate / disactivate the automatic generation of directory listings by the server.
This directive can be used to deactivate all filter rules on a per directory basis (in .nwaccess files).
Sets a filter rule which is applied to all files matching the mime pattern
(most often text/html, but may be just */*).
The second argument to this directive is the name of one of the internal filters which you may pass additional arguments.
The most powerful filter is 'pipe' which passes the current content through an external filter program given as argument to it ("/usr/bin/tidy" for example). The 'unchunk' filter is very often required to be listed before the first 'pipe'. Other filters are currently 'null', 'shrink', 'downcase' and 'wap' (you need to load »mod_html_filters« to make them available).
This directive is to be used in .nwaccess files to enforce one of the nanoweb parsers (like CGI, static or SSI) to handle all files. So if you want to ensure, that none of the files in a directory is activated as CGI script, you could put the line "ForceHandler=static" into the .nwaccess file.
Default group nanoweb will run as (defaults to »www-data«).
This directive allows you to enable/disable mod_gzip for each virtual host
or different directories (in .nwaccass files).
Chunked CGI output will not get compressed as this often leads to poor results and is not supported by some browsers. You can however enforce compression of such content streams by setting this directive to 2 - A much better solution is however to apply the "unchunk" Filter, which will help mod_gzip. For PHP scripts you may wish to enable automagic output compression via php.ini rather.
This options tells mod_gzip how much time to spend on compression. A value
of 1 leads to very fast processing, 9 means maximum compression, which
is of course very time consuming. So you probably want to have some value in
between to achieve good compression in a justifiable amount of time.
BTW, 0 means no compression, but this is stupid; see the above configuration directive for this.
Max compression ratio (percentage of original size). If it is higher, the content is likely to be already compressed and does not need a recompression.
Use reverse-DNS to put hostnames instead of IP addresses into server log.
Note that using DNS slows down nanoweb a bit, so if you need some more speed you should think about disabling this feature, as you can get the names of the hosts which requested pages from your server later too by evaluating the server logs with appropriate software.
Using the value "logger" in this directive tells nanoweb to delay DNS lookups and let these be done by the logger processes after the page already has been delivered. This is a speed improvement, because there is actually no need to do the slow DNS query for the requesting host in the main server; some modules may however depend on it, so you can set this directive to "server" for the old behaviour.
Setting this directive to 1 will make nanoweb ignore .xxx files. This is for security reasons (not allowing clients to request .nwaccess or .nwauth files for example). You shouldn't change this unless you know exactly what you're doing.
This directive reads another (swapped out) configuration file into the current one. In standard installations the main configuration file is /etc/nanoweb/nanoweb.conf which additionally loads modules.conf and vhosts.conf using this directive.
Maximum number of http requests per connection (A client may get more than one file after it opened a connection to nanoweb, to speed up downloading of multiple files). To disable the keep-alive feature of nanoweb just set this value to 0.
Content-negotiation occurs if the requested file does not exist, but some variants can be found instead. If the client doesn't specifies a preferred language this setting will come into use.
Specify the IP address the server shall listen to for incoming connections.
Specify the TCP port number the server shall listen on for incoming
connections. 80 is the default for web servers, but 8080 is also very
common and you must use this to run nanoweb if you're not root (as ports
below 1024 are reserved for system daemons).
Starting with version 2.0 you can make Nanoweb listen on more than one TCP port, if you give this directive multiple times:
ListenPort = 80 ListenPort = 8080 ListenPort = 12080
Maximum number of queued connections.
Defines the maximum load average after which the server will return an error and won't serve documents until load goes down.
Per dafault nanoweb stops serving any further requests if the load limit is reached, but instead of just sending an error message response nanoweb is also able to redirect the request to another server (and thus probably all further queries from the requesting client).
This directive sets the http response code for error messages that are send back when the the server load gets too high. (rarely needed)
Defines a custom error message that is send if your server gets too slow. Besides the common nanoweb server vars, you can use %CUR_LOAD and %MAX_LOAD in the message text.
If enabled with the LoadLimitAction directive,
clients are redirected to the server given by this directive as soon
as the load limit is reached.
Please note that %DOCUMENT_ROOT may be useful together with %REQUEST_URI if your fallback server holds backups of different sites.
This directive is usually found in /etc/nanoweb/modules.conf
(or C:\nanoweb\modules.conf on Windows),
which is included into the main server configuration
file. It registers the specified module to the server core in the startup
sequence of nanoweb.
You don't need anymore to specify the full pathname of the PHP source file containing the module if you set ModulesDir accordingly.
Nanoweb supports themes to be applied to server generated response messages.
This directive loads a template file on server startup, which then can be
activated on a per-directory basis using the
ServerTheme directive. You can give the full path
to the theme file; but if you set the default configuration path using
the ConfigDir directive you can leave that out,
as theme files are usually located there, too.
Theme files can be loaded from inside a .nwaccess configuration file, too.
For more information about this, please see the manual section about Themes.
Specifies the file where nanoweb notes all served requests. This file will
contain one line for each requested file with informations on where the
request origined, when it occoured, and if it succeded (file found) beside
with some other notes.
The format of this file is the same as with apache and often referred to as »common« or »combined« »log file format«, thus you can run some evalution program like "awstats" on it to generate a monthly traffic report for your site.
You can leave out the absolute path name, if a default path is already set by LogDir.
Logger Process support. If enabled, nanoweb will spawn logger processes at launch and all logging will be done by them. This can help if you are using User or Group directives or simply want more performance from the server. Set to the number of loggers you want, or 0 to disable.
User the nanoweb loggers will run as.
Group the nanoweb loggers will run as.
LogHitsToConsole replaced this directive.
Logs every hit to standard output. Of course this requires you not to run nanoweb in daemon mode via nanoctl start, you need to run nanoweb.php from your vterminal/shell. Additionally the --quiet option given to nanoweb.php allows you to disable this on the fly (this is neccessary for inetd mode for example), so you don't need to change the configuration files.
Logs hits additionally to the standard server log, which normally wouldn't contain these (access.log used to contain hits).
Sets the format used for log files (see Log) output. »common« is the NCSA Common Log Format (also known as »clf«), »common-with-vhost« is the same as CLF, but with support for multiple hosts, »combined« is the extended CLF format used by Apache, which also reports user agent and referer informations.
Limits the size of the request body (the content of POST and PUT request),
Sets the maximum length of the request URIs nanoweb will accept. Usually the size of URLs is already limited at client side (IE for example: 1024 characters) but for security reasons you may wish to trim it here.
Limits the number of subservers (child processes) Nanoweb will create at maximum. A setting of 0 permits unlimited server processes.
This directive tells mod_meta how to fetch the http header fields given in the <META HTTP-EQUIV> tags in .html files. The default fast method is only suitable if your html files are very well formed, the regex method is often more reliable and fetches non-http meta information as well, but is therefor much slower. The no disables opening the .html files for scanning, so only the headers found in every files companion *.meta file are attached to http answers. mod_meta gets fully disabled only by setting this directive to 0.
Defines a 'httpd user space' CGI script that handles HTTP requests with non-standard methods. The given script is called like a standard handler (as in mod_misc_filters) with the requested file given in $PATH_INFO. The path to the handler script must be absolute to the docroot.
The mispell module handles incorrectly
entered URLs. If a file with the given name cannot be found on the server
this module tries to figure out the correct name by checking the sounding of the
filename against that of the files located on the server.
This directives says what to do if it finds the correct name for the requested file. Default is »advice« which sends back an error page to the user, presenting him the corrected URL. You can also specify »redirect« which will send back a HTTP redirection response to the browser, so the users browser automagically re-requests with the corrected URL.
Reads in the system-wide available MIME-types database from given file, and thus makes nanoweb recognize hundreds of typical file extensions.
This directive specifies the default location of the nanoweb extension
modules, so you don't need to give the full path name in
LoadModule directives. You must append
a slash to the directory name.
You may give this directive multiple times, nanoweb will use the module it finds first in one of the given paths. Due to PHP's fopen wrapper mechanism you could also give a http:// address here to always have the most current module versions available - but this is still under devolopment (and would be mentioned in one of your configuration files then).
This directive enables / disables mod_multiviews, so you can use the negotiation features only in directories you actually need them.
The mysqllog module saves request hits to a database instead of saving it to files as mod_stdlog does. So you need to give a hostname here, where a MySQL database server runs on.
Sets the name of the MySQL database for request logging.
You need an account for access to a MySQL database server.
You need an account for access to a MySQL database server.
This gives additional preferences that will help mod_multiviews in selecting the "best" variant when the browser didn't send an Accept: header. Note that you can mix file extensions (without dot!) and mime types; quality-values help a lot.
Defines which script interpreter must be run to parse files with a given
Valid parser TYPEs are defined by the modules you loaded and the parameters are type dependant. CGI is the most commonly used parser type and requires the absolute filename of the interpreter to execute. FCGI argument is the FastCGI server socket, while SSI never requires any parameter.
See also the pages for mod_cgi mod_fcgi and mod_include and this example:
ParseExt = php CGI /usr/local/bin/php ParseExt = awk CGI /usr/bin/awk $FILENAME ParseExt = js CGI /usr/bin/js $FILENAME
This directive gives CGI extensions that should be checked for if a file
with a pathinfo is requested but no file extension is there. You should use
multiple PathInfoTryExt lines, one for each CGI extension you would like to
See also AllowPathInfo
This directive gives a filename where nanoweb logs which pid (=process id, on Unix systems) it runs as.
The proxy module generates an additonal seperate log file, as you probably don't want the main server log to be intermixed with request informations for files that are not located in your servers docroot.
Use this directive to allow hosts or all hosts from a network (with an abbreviated IP address range) to access the www via the nanoweb internal proxy server. Hosts not listed here cannot make use of the nanoweb proxy.
The proxy module of course needs a dedicated directory where it can save all the files that were requested in the past, so that they do not need to be fetched a second time from the real server if they get requested another time.
Cached pages get invalid (need to be refetched) after an amount of time specified with this directive, the value is in seconds.
If you want some servers not to be accessible through your proxy you just need to list them in a text file (one server name per line) which you specify here; so access to such servers will be forbidden.
The proxy module can additonally filter popups out of delivered pages, if they match one of the regexs listed in the specified file.
This directive enables you to filter images. Just put a line with the filename or URL into the file »images.txt« for every image, that you do not want to be delivered through the proxy.
In a context where RefererCheck has been activated, this allows requests refered by the defined URL (or part of URL) to be processed as if they were coming from the local site
When activated, nanoweb will answer every request which referer is nonempty and differs from the ServerName with a 403 Forbidden error page.
Other exceptions than the ServerName can be set with the RefererAllow directive.
If a URL is rewritten via a RewriteRule or because of some content
negotiation, this setting influences if the rewritten URL (or if unset
the requested one) shall be presented in error messages that may occur.
Enabling this probably leads to trouble if you also run mod_proxy.
Mappes the requested file name to real one by rewriting with usage of regular expressions pattern matching. Please refer to the modules` documentation for a short explanation.
Does additional regex tests that trigger if the following RewriteRule gets executed (on success of this RewriteCond).
Nanoweb will wait specified amount of time for receival of the actual http request after a TCP/IP connection was established, before it closes this connection again to not prevent others from correctly connecting.
Admin mail address present in most of the server generated pages (errors, status, directory listings).
As there can be only one default ServerName, you
may use this directive to let nanoweb listen to additional DNS names (just
give multiple lines here).
You can optionally give a port number after the server name.
If you'd like to annoy possible attackers to your webserver you can give
them false informations on which webserver you are actually running using
You have to enable this behaviour by additionally changing the ServerSignature directive to 'fake'.
The ServerLog directive may be given multiple times, to let nanoweb create separate
files where runtime informational messages are written to. Having multiple
of them makes sense, because the second parameter (logmode) to this directive allows
you to filter parts of the possible server messages.
The internally generated server messages are divided into the classes: error, notice warning, debug, and hit. You can even say all or default here, or give multiple modes by connecting them with the plus sign, like "error+warning+hit" for example.
ServerLog = /var/log/nanoweb/server.log default+hit ServerLog = /dev/tty9 all ServerLog = /var/log/nanoweb/error.log error+warning ServerLog = /var/log/nanoweb/server.debug notice+warning
You may wish to run nanoweb in standalone mode in most cases. This is,
the nanoweb server is started to stay resident in memory and waiting for
Another variant of using the nanoweb server is by invoking it via inetd - the »internet superserver« used on many UNIX systems to startup internet services on oncoming requests, which leads to some overhead and is slower in general as one copy of nanoweb.php needs to be executed for each requested file.
Default DNS-name of this server. You may append a port number to the name.
Webservers usually reveal their name and version in the HTTP answer headers,
but you can restrict nanoweb in how much information it gives away. Per
default (full) it tells "nanoweb/NW_VERSION OPERATING_SYS PHP_VERSION",
with this directive set to os it leaves out the PHP version,
min still includes the nanoweb version but prod only tells
"nanoweb" (the product name).
Finally you can disable the output fully with off. But much more funny it is to give clients wrong information instead using fake and the ServerFakeSignature.
This directive selects one of the theme templates (loaded using LoadTheme) to be applied to server generated responses like error messages or directory listings.
This is used to force nanoweb in single process mode (which is usually slower than the default process forking mode). You should uncomment it when running in win32 or if your php binary was not compiled with pcntl.
This gives an IP of a host allowed to consult the server status through /server-status. Multiple lines with this directive may be specified; an incomplete IP address can be used to allow a range of host addresses to be granted access.
Limits the chunk size of static files to 1M per default, because nanoweb cannot keep files with unlimited size in memory. You may in fact speed up requests (for some cases) if you lower this value.
Disallows http clients to retrieve only parts of a resource/file, when set to 1. Usually the partial GET is used by clients like 'wget' to resume a stopped/unfinished previous download attemp.
This specifies where nanoweb shall put temporary files into, so this directory should be read+writeable by the user/group nanoweb will run as.
This directive was renamed to TempDir in 2.1.1 to match general naming syntax.
Default user nanoweb will run as (the user name is »www-data« per dafault).
This is the subpath which nanoweb will look for in users homedir
when an URL like http://webserver/~user is requested.
This directive can be set in the global section or for each virtual host.
Commands specified with such directives are executed, if a worm is detected.
Please refer to the file README.mod_worms (in the docs/ directory of the unpacked tarball) as mod_worms requires additional preparation to work.
Text that is sent to the offending IIS server console.
This directive enables writing files via HTTP for clients that support the
PUT request method.
You shouldn't enable this setting in the main configuration file or one of the vhost sections; it is often a good idea to allows this type of access only for selected directories (in a .nwaccess file), even if the PUT module only can write to files and directories that have the world-writable bit set or that are owned by the default Nanoweb user/group.