HTTP(s) (im_http)

This module listens for HTTP or HTTPS connections on the address and port defined by the ListenAddr directive. It accepts HTTP requests that include content and use methods such as POST, PUT, or PATCH. However, the module does not distinguish between the different HTTP methods. When using compression, im_http accepts requests with the Content-Encoding header set to identity, deflate, and gzip.

The log event must be included in the request body as raw data and will be available in the $raw_event field. The request must include the Content-Length header indicating the size of the event, and the module will read the number of bytes specified by this header. The module does not close connections while receiving valid requests to operate in Keep-Alive mode. It responds with HTTP/1.1 200 OK to each valid request. This acknowledgment ensures reliable message delivery.

im_http contains a partial HTTP/1.1 implementation designed to work with the om_http module. It has received limited interoperability testing, and thus, using it with other third-party clients may not have the expected results.

To examine the supported platforms, see the list of installation packages.

Configuration

The im_http module accepts the following directives in addition to the common module directives.

HTTPS directives

The following directives configure secure data transfer via HTTPS.

HTTPSAllowExpired

Specifies if the connection should be allowed with an expired certificate. If set to TRUE, the remote host will be able to connect with an expired certificate. The default is FALSE: the certificate must not be expired. This directive is only valid if HTTPSRequireCert is set to TRUE.

HTTPSAllowHostnameValidation

Specifies if the certificate FQDN should be validated against the server hostname or not. If set to TRUE, the connection will only be allowed if the certificate FQDN corresponds to the server hostname. The default value is FALSE: the remote server hostname is not validated.

HTTPSAllowUntrusted

Specifies if the connection should be allowed without certificate verification. If set to TRUE, the connection will be allowed even if the remote host presents an unknown or self-signed certificate. The default value is FALSE: the remote host must present a trusted certificate.

HTTPSCADir

The path to a directory containing certificate authority (CA) certificates. These certificates will be used to verify the certificate presented by the remote host. The certificate files must be named using the OpenSSL hashed format, i.e. the hash of the certificate followed by .0, .1 etc. To find the hash of a certificate using OpenSSL:

$ openssl x509 -hash -noout -in ca.crt

For example, if the certificate hash is e2f14e4a, then the certificate filename should be e2f14e4a.0. If there is another certificate with the same hash then it should be named e2f14e4a.1 and so on.

A remote host’s self-signed certificate (which is not signed by a CA) can also be trusted by including a copy of the certificate in this directory.

The default operating system root certificate store will be used if this directive is not specified. Unix-like operating systems commonly store root certificates in /etc/ssl/certs. Windows operating systems use the Windows Certificate Store, while macOS uses the Keychain Access Application as the default certificate store. See Certification Authority (CA) certificates in the NXLog Platform User Guide for more information on using this directive.

In addition, Microsoft’s PKI repository contains root certificates for Microsoft services.

HTTPSCAFile

The path of the certificate authority (CA) certificate that will be used to verify the certificate presented by the remote host. A remote host’s self-signed certificate (which is not signed by a CA) can be trusted by specifying the remote host certificate itself. In case of certificates signed by an intermediate CA, the certificate specified must contain the complete certificate chain (certificate bundle).

HTTPSCAPattern

You can use this directive on Windows and macOS to specify a PCRE2-compliant regular expression for locating a suitable Certificate Authority (CA) certificate. On Windows, the certificate must be in the Windows Certificate Store. On macOS, the certificate must be in the system keychain.

The pattern must be in the format "SUBJECT=, CN=; DN=CN=, O=, OU=, L=, ST=, C=; SAN=". For example:

HTTPSCAPattern    'DN=CN=SERVER01\.example\.com.*?SAN=DNS:SERVER01\.example\.com'

The above configuration will result in the following logging in the NXLog Agent log file:

matching pattern [DN=CN=SERVER01\.example\.com;.*?SAN=DNS:SERVER01\.example\.com] to certificate [SUBJECT=US, ClientState, ClientCity, ClientCompany, ClientUnit, SERVER01.example.com, CN=SERVER01.example.com; DN=CN=SERVER01.example.com, O=ClientCompany, OU=ClientUnit, L=ClientCity, ST=ClientState, C=US; SAN=DNS:SERVER01.example.com; DNS:www.SERVER01.example.com; IP:127.0.0.3; ]

If the pattern matches multiple certificates, NXLog Agent will use the first one. If the certificate changes, you must restart NXLog Agent for it to start using the new certificate.

This directive is mutually exclusive with the HTTPSCAThumbprint directive.

On macOS, keychain behavior and certificate requirements depend on your operating system version. Newer versions enforce stricter rules on certificate algorithms and key sizes. A certificate and private key may be imported into the keychain, yet remain unusable if they do not comply with the current macOS requirements. After importing, you may also need to give NXLog Agent permission to use the certificate in the keychain.

Check your operating system documentation to find out which certificates are trusted and valid. See also Apple’s certificate requirements for macOS.

HTTPSCAThumbprint

This optional directive, supported on Windows and macOS, specifies the thumbprint of the certificate authority (CA) certificate that will be used to verify the certificate presented by the remote host.

On Windows, the hexadecimal fingerprint string can be copied from Windows Certificate Manager (certmgr.msc). The certificate must be added to a Windows certificate store that is accessible by NXLog Agent.

On macOS, the hexadecimal fingerprint string can be obtained from the Keychain Access application. The certificate must be accessible in the system keychain.

Whitespaces and colons are automatically removed.

This directive is mutually exclusive with the HTTPSCADir and HTTPSCAFile directives.

Check your operating system documentation to find out which certificates are trusted and valid. See also Apple’s certificate requirements for macOS.

HTTPSCertFile

The path of the certificate file that will be presented to the remote host during the HTTPS handshake.

HTTPSCertKeyFile

The path of the private key file that was used to generate the certificate specified by the HTTPSCertFile directive. This is used for the HTTPS handshake.

HTTPSCertPattern

This optional directive, supported on Windows and macOS, defines a pattern for identifying a corresponding certificate and its private key. The pattern must follow PCRE2 rules and use the format "SUBJECT=, CN=, DN=, SAN=" where DN is "CN=, O=, OU=, L=, ST=, C=". On Windows, the certificate must be imported in PFX format into the Local Computer\Personal certificate store for NXLog Agent to locate it. On macOS, the certificate and its associated private key must be present in the System Keychain (/Library/Keychains/System.keychain). During configuration, this directive is resolved into the corresponding CertThumbprint value. The first found thumbprint will be chosen if multiple certificates match the pattern. We recommend ensuring that the used certificate storage is well-maintained for optimal performance. This feature is not dynamic; the agent must be restarted if the certificate changes. This directive is mutually exclusive with the HTTPSCertThumbprint directive.

Configuration examples:

HTTPSCertPattern    $hostname + 'Cert'

HTTPSCertPattern    DN=CN=Client\.example\.com;.*?SAN=DNS:Client\.example\.com

A normal log output example would look like as follows:

matching pattern [DN=CN=Client\.example\.com;.*?SAN=DNS:Client\.example\.com] to certificate [SUBJECT=US, ClientState, ClientCity, ClientCompany, ClientUnit, Client.example.com, CN=Client.example.com; DN=CN=Client.example.com, O=ClientCompany, OU=ClientUnit, L=ClientCity, ST=ClientState, C=US; SAN=DNS:Client.example.com; DNS:www.Client.example.com; IP:127.0.0.3; ]

Check your operating system documentation to find out which certificates are trusted and valid. See also Apple’s certificate requirements for macOS.

HTTPSCertThumbprint

This optional directive, supported on Windows and macOS, specifies the thumbprint of the certificate that will be presented to the remote host during the HTTPS handshake. Whitespaces and colons are automatically removed.

On Windows, the hexadecimal fingerprint string can be copied from Windows Certificate Manager (certmgr.msc). The certificate must be imported to the Local Computer\Personal certificate store in PFX format for NXLog Agent to find it. Run the following command to create a PFX file from the certificate and private key using OpenSSL:

$ openssl pkcs12 -export -out server.pfx -inkey server.key -in server.pem

When the global directive UseCNGCertificates is set to FALSE the private key associated with the certificate must be exportable.

If you generate the certificate request using Windows Certificate Manager, enable the Make private key exportable option from the certificate properties.
If you import the certificate with the Windows Certificate Import Wizard, make sure that the Mark this key as exportable option is enabled.
If you migrate the certificate and associated private key from one Windows machine to another, select Yes, export the private key when exporting from the source machine.

On the contrary, when the global directive UseCNGCertificates is set to TRUE the private key associated with the certificate does not have to be exportable. In cases like TPM modules, the private key is always nonexportable.

On macOS, the hexadecimal fingerprint string can be obtained from the Keychain Access application. The certificate and its associated private key must be present in the System Keychain (/Library/Keychains/System.keychain).

The usage of the directive is the same in all cases:

HTTPSCertThumbprint    7c2cc5a5fb59d4f46082a510e74df17da95e2152

Check your operating system documentation to find out which certificates are trusted and valid. See also Apple’s certificate requirements for macOS.

HTTPSCRLDir

The path to a directory containing certificate revocation list (CRL) files. These CRL files will be used to check for certificates that were revoked and should no longer be accepted. The files must be named using the OpenSSL hashed format, i.e. the hash of the issuer followed by .r0, .r1 etc. To find the hash of the issuer of a CRL file using OpenSSL:

$ openssl crl -hash -noout -in crl.pem

For example if the hash is e2f14e4a, then the filename should be e2f14e4a.r0. If there is another file with the same hash then it should be named e2f14e4a.r1 and so on.

HTTPSCRLFile

The path of the certificate revocation list (CRL) which will be used to check for certificates that have been revoked and should no longer be accepted. Example to generate a CRL file using OpenSSL:

$ openssl ca -gencrl -out crl.pem

HTTPSDHFile

This optional directive specifies a file with dh-parameters for Diffie-Hellman key exchange. These parameters can be generated with dhparam(1ssl). If no directive is specified, default parameters will be used. See OpenSSL Wiki for further details.

HTTPSKeyPass

The passphrase of the private key specified by the HTTPSCertKeyFile directive. A passphrase is required when the private key is encrypted. Example to generate a private key with Triple DES encryption using OpenSSL:

$ openssl genrsa -des3 -out server.key 2048

This directive is not needed for passwordless private keys.

HTTPSLoadCertificateChains

If set to TRUE, try to load higher-level certificates from the referenced PEM file which may contain only one certificate or the whole chain. The default value is FALSE: certificates will be instead loaded from the operating system certification storage.

This directive is only supported on Windows.

HTTPSRequireCert

Specifies if the remote HTTPS host must present a certificate. If set to TRUE and there is no certificate presented during the connection handshake, the connection will be refused. The default value is TRUE: each connection must use a certificate.

HTTPSSearchAllCertStores

This optional directive, if set to TRUE, enables the loading of all available Windows certificates into NXLog Agent, for use during remote certificate verification. Any required certificates must be added to a Windows certificate store that NXLog Agent can access. This directive is mutually exclusive with the HTTPSCAThumbprint, HTTPSCADir and HTTPSCAFile directives.

This directive is only supported on Windows.

HTTPSSSLCipher

This optional directive can be used to set the permitted SSL cipher list, overriding the default. Use the format described in the ciphers(1ssl) man page. For example specify RSA:!COMPLEMENTOFALL to include all ciphers with RSA authentication but leave out ciphers without encryption.

If RSA or DSA ciphers with Diffie-Hellman key exchange are used, DHFile can be set for specifying custom dh-parameters.

HTTPSSSLCiphersuites

This optional directive can be used to set the permitted cipher list for TLSv1.3. Use the same format as in the HTTPSSSLCipher directive. Refer to the OpenSSL documentation for a list of valid TLS v1.3 cipher suites. The default value is:

TLS_AES_256_GCM_SHA384:TLS_CHACHA20_POLY1305_SHA256:TLS_AES_128_GCM_SHA256

HTTPSSSLCompression

If set to TRUE, the module uses TLS compression when sending data over the network. The compression mechanism is based on the zlib compression library. The default is FALSE, meaning TLS compression is not used.

Some NXLog Agent packages for Linux, such as Debian, utilize the OpenSSL library provided by the OS, which may not support zlib compression. The module will emit a warning on startup if compression support is missing. The generic deb/rpm packages are bundled with a zlib-enabled libssl library.

HTTPSSSLProtocol

This directive can be used to set the allowed SSL/TLS protocol(s). It takes a comma-separated list of values which can be any of the following: SSLv2, SSLv3, TLSv1, TLSv1.1, TLSv1.2 and TLSv1.3. By default, the TLSv1.2 and TLSv1.3 protocols are allowed. Note that the OpenSSL library shipped by Linux distributions may not support SSLv2 and SSLv3, and these will not work even if enabled with this directive.

HTTPSTLSCertLog

Use this directive to log the remote host’s certificate details during TLS/SSL handshake, to help with debugging. Setting this directive to TRUE logs, if available, the Subject, Issuer, NotBefore and NotAfter fields to the NXLog Agent log file. The default is FALSE.

This directive depends on the values of the HTTPSAllowUntrusted and HTTPSRequireCert directives:

HTTPSAllowUntrusted HTTPSRequireCert Outcome

FALSE

Output depends on HTTPSTLSCertLog value

FALSE

TRUE

Output depends on HTTPSTLSCertLog value

TRUE

FALSE

No output regardless of HTTPSTLSCertLog value

TRUE

Output depends on HTTPSTLSCertLog value

Sigalgs

The signature algorithm parameter that is being sent to the Windows SSL library. Allowed values depend on the available encryption providers.

This directive is only supported on Windows.

SNI

This optional directive specifies the hostname used for Server Name Indication (SNI) in HTTPS mode.

UseCNGCertificates

If set to TRUE, the module uses the Windows Cryptography API: Next Generation (CNG) to access the private keys associated with certificates identified by a thumbprint.

This directive is only supported on Windows.

Optional directives

ListenAddr

Set this directive to the IP address or DNS hostname on which the module accepts connections. The default is localhost, which restricts access to the local machine and prevents connections from remote hosts.

Append the port number to the address using a colon separator (host:port). The default port is 80. For IPv6 addresses, enclose the address in square brackets ([addr]:port). Use 0.0.0.0:80 for IPv4 or [::]:80 for IPv6 to accept connections on all interfaces. NXLog Agent requires additional privileges to bind to ports below 1024 on Linux. See Granting additional Linux capabilities for more information.

You can specify this directive multiple times to accept connections on multiple addresses or ports within a single module instance. Set IPv4 and IPv6 addresses as separate entries. If you use a DNS name, keep the number of addresses or CNAMEs below 16 to avoid errors caused by DNS response size limits.

AllowIP

Set this directive to restrict incoming connections to specific IP addresses or networks. You can specify this directive multiple times to allow multiple IPs or networks. AllowIP has no effect unless ListenAddr is configured.

If AllowIP is not set, the module falls back to BlockIP to deny specific connections. If neither AllowIP nor BlockIP is set, the module does not restrict incoming connections.

The following IP address formats may be used:

0.0.0.0 (IPv4 address)
0.0.0.0/32 (IPv4 network with subnet bits)
0.0.0.0/0.0.0.0 (IPv4 network with subnet address)
aa::1 (IPv6 address)
aa::12/64 (IPv6 network with subnet bits)

BlockIP

Set this directive to deny incoming connections from specific IP addresses or networks. You can specify this directive multiple times to deny multiple IPs or networks. BlockIP has no effect unless ListenAddr is configured.

If BlockIP is not set, the module falls back to AllowIP to restrict incoming connections. If neither BlockIP nor AllowIP is set, the module does not restrict incoming connections.

The following IP address formats may be used:

0.0.0.0 (IPv4 address)
0.0.0.0/32 (IPv4 network with subnet bits)
0.0.0.0/0.0.0.0 (IPv4 network with subnet address)
aa::1 (IPv6 address)
aa::12/64 (IPv6 network with subnet bits)

MaxConnections

Set this directive to limit the number of concurrent active connections for a listening TCP socket. The default is 4294967295 (unlimited). When the limit is reached, the module rejects new connections and logs an error:

Number of allowed active connections(10) reached: 10. Refusing connection from 127.0.0.1

ConnectionIdleTimeout

Set this directive to close TCP connections that have been idle for longer than the specified number of seconds. The minimum value is 15 seconds. If this directive is not set, the module keeps idle TCP connections open indefinitely.

ExclusiveAddrUse

Set this directive to TRUE to prevent other processes or module instances from binding to the same port.

The default is FALSE; multiple module instances can bind to the same port.

This directive is only supported on Windows.

ReuseAddr

Set this directive to TRUE to allow the module instance to bind to a port that is already in use.

The default is TRUE; multiple module instances can listen on the same port and process data simultaneously.

ReusePort

Set this directive to TRUE to allow multiple module instances to listen on the same port. Each instance runs in a separate thread, allowing NXLog Agent to process incoming logs simultaneously.

The default is FALSE; multiple module instances cannot bind to the same port.

This directive is not supported on Windows.

Functions

The following functions are exported by im_http.

type: string get_abs_path(): Returns the absolute path of the URL.
type: string get_param(type: string name): Returns the value of the specific URL parameter.
type: string get_rel_path(): Returns the relative path of the URL.
type: string get_url(): Returns the full URL.
type: hash parse_header(): Returns the HTTP headers for the current record (last parsed http request) as a hash value.
type: string parse_header(type: string key): From the current record, this returns the HTTP header value of the key.

Procedures

The following procedures are exported by im_http.

parse_url_params();: Generates fields from URL query parameters.

Fields

The following fields are created by im_http.

$raw_event (type: string): The content received in the POST request.
$MessageSourceAddress (type: ipaddr): The IP address of the remote host.

Examples

Example 1. Receiving logs over HTTPS

This configuration listens for HTTPS connections from localhost. Received log messages are written to a file.

nxlog.conf

<Input http>
    Module              im_http
    ListenAddr          0.0.0.0:8888
    HTTPSCertFile       %CERTDIR%/server-cert.pem
    HTTPSCertKeyFile    %CERTDIR%/server-key.pem
    HTTPSCAFile         %CERTDIR%/ca.pem
    HTTPSRequireCert    TRUE
    HTTPSAllowUntrusted FALSE
</Input>

Example 2. Receiving logs over HTTPS using certificate thumbprints

This configuration uses the HTTPSCAThumbprint and HTTPSCertThumbprint directives for the verification of the Certificate Authority and the SSL handshake respectively.

nxlog.conf

<Input in_https>
    Module                 im_http
    ListenAddr             0.0.0.0:443
    HTTPSCAThumbprint      c2c902f736d39d37fd65c458afe0180ea799e443
    HTTPSCertThumbprint    7c2cc5a5fb59d4f46082a510e74df17da95e2152
    HTTPSSSLProtocol       TLSv1.2
</Input>

Example 3. Sending data to im_http over HTTPS

This script uses the curl utility to send data to an im_http instance. The log data is passed as an argument when executing the script.

The URL value needs to be replaced with a valid hostname according to the value specified by the ListenAddr directive.

The PORT value needs to be replaced with the actual port that im_http has been configured to listen on.

The USESSL value needs to be updated with the path to the client certificate, client certificate key, and certification authority files respectively. In this example, the files are located in a folder named keys located in the same path as the script.

im-http-curl-request.sh

URL=https://<hostname>
PORT=<port>
USESSL="--cert keys/client-cert.pem --key keys/client-key.pem --cacert keys/ca.pem"

BYTELEN=`printf "%s" "$1" | wc -c`

curl -v $USESSL -H "Content-Type:plain/text" -H "Content-Length:${BYTELEN}" -d "$1" $URL:$PORT

The script is executed by running the following command and passing the log data to it as an argument:

# path/to/im-http-curl-request.sh "My log line"

This bash script and other examples can be found in the NXLog public repository on GitLab.

Example 4. Parsing values from incoming HTTP headers

This configuration waits for HTTP packets and parses the incoming headers, then logs the requested information. It extracts the header values, such as the User-Agent and the Content-Length, first by requesting the specific field directly, then by parsing the entire header and extracting the value from the hash. Both calls use compound values, but they can be used with simple values, depending on the scope of the implementation.

nxlog.conf

<Input ihttp>
Module          im_http
ListenAddr      0.0.0.0:6060
<Exec>
    $$agent = parse_header("User-Agent");
    log_info("User Agent:" + $$agent);

    # retrieving a hash with the entire header
    $$entire_header = parse_header();
    log_info("Data segment size in Bytes:" + $$entire_header('Content-Length'));
    log_info("Entire header parsed:" + $$entire_header);
</Exec>
</Input>