The Python destination allows you to write your own destination in Python. You can import external Python modules to process the messages, and send them to other services or servers. Since many services have a Python library, the Python destination makes integrating syslog-ng OSE very easy and quick.
The following points apply to using Python blocks in syslog-ng OSE in general:
-
Python parsers and template functions are available in syslog-ng OSE version 3.10 and later.
Python destinations and sources are available in syslog-ng OSE version 3.18 and later.
-
Supported Python versions: 2.7 and 3.4+ (if you are using pre-built binaries, check the dependencies of the package to find out which Python version it was compiled with).
-
The Python block must be a top-level block in the syslog-ng OSE configuration file.
-
If you store the Python code in a separate Python file and only include it in the syslog-ng OSE configuration file, make sure that the PYTHON_PATH environment variable includes the path to the Python file, and export the PYTHON_PATH environment variable. For example, if you start syslog-ng OSE manually from a terminal and you store your Python files in the /opt/syslog-ng/etc directory, use the following command: export PYTHONPATH=/opt/syslog-ng/etc
In production, when syslog-ng OSE starts on boot, you must configure your startup script to include the Python path. The exact method depends on your operating system. For recent Red Hat Enterprise Linux, Fedora, and CentOS distributions that use systemd, the systemctl command sources the /etc/sysconfig/syslog-ng file before starting syslog-ng OSE. (On openSUSE and SLES, /etc/sysconfig/syslog file.) Append the following line to the end of this file: PYTHONPATH="<path-to-your-python-file>", for example, PYTHONPATH="/opt/syslog-ng/etc"
-
The Python object is initiated every time when syslog-ng OSE is started or reloaded.
|
|
Caution:
If you reload syslog-ng OSE, existing Python objects are destroyed, therefore the context and state information of Python blocks is lost. Log rotation and updating the configuration of syslog-ng OSE typically involves a reload. |
-
The Python block can contain multiple Python functions.
-
Using Python code in syslog-ng OSE can significantly decrease the performance of syslog-ng OSE, especially if the Python code is slow. In general, the features of syslog-ng OSE are implemented in C, and are faster than implementations of the same or similar features in Python.
-
Validate and lint the Python code before using it. The syslog-ng OSE application does not do any of this.
-
Python error messages are available in the internal() source of syslog-ng OSE.
-
You can access the name-value pairs of syslog-ng OSE directly through a message object or a dictionary.
-
To help debugging and troubleshooting your Python code, you can send log messages to the internal() source of syslog-ng OSE. For details, see Logging from your Python code.
NOTE: Starting with 3.26, syslog-ng OSE assigns a persist name to Python sources and destinations. The persist name is generated from the class name. If you want to use the same Python class multiple times in your syslog-ng OSE configuration, add a unique persist-name() to each source or destination, otherwise syslog-ng OSE will not start. For example:
log {
source { python(class(PyNetworkSource) options("port" "8080") persist-name("<unique-string>); };
source { python(class(PyNetworkSource) options("port" "8081")); };
};
Alternatively, you can include the following line in the Python package: @staticmethod generate_persist_name. For example:
from syslogng import LogSource
class PyNetworSource(LogSource):
@staticmethod
def generate_persist_name(options):
return options["port"]
def run(self):
pass
def request_exit(self):
pass
Declaration:
Python destinations consist of two parts. The first is a syslog-ng OSE destination object that you define in your syslog-ng OSE configuration and use in the log path. This object references a Python class, which is the second part of the Python destination. The Python class processes the log messages it receives, and can do virtually anything that you can code in Python. You can either embed the Python class into your syslog-ng OSE configuration file, or store it in an external Python file.
destination <name_of_the_python_destination>{
python(
class("<name_of_the_python_class_executed_by_the_destination>")
);
};
python {
class <name_of_the_python_class_executed_by_the_destination>(object):
def open(self):
"""Open a connection to the target service
Should return False if opening fails"""
return True
def close(self):
"""Close the connection to the target service"""
pass
def is_opened(self):
"""Check if the connection to the target is able to receive messages"""
return True
def init(self, options):
"""This method is called at initialization time
Should return false if initialization fails"""
return True
def deinit(self):
"""This method is called at deinitialization time"""
pass
def send(self, msg):
"""Send a message to the target service
It should return True to indicate success. False will suspend the
destination for a period specified by the time-reopen() option."""
return True
def flush(self):
"""Flush the queued messages"""
pass
};
Methods of the python() destination
init(self, options) method (optional)
The syslog-ng OSE application initializes Python objects every time when it is started or reloaded. The init method is executed as part of the initialization. You can perform any initialization steps that are necessary for your source to work.
|
|
Caution:
If you reload syslog-ng OSE, existing Python objects are destroyed, therefore the context and state information of Python blocks is lost. Log rotation and updating the configuration of syslog-ng OSE typically involves a reload. |
When this method returns with False, syslog-ng OSE does not start. It can be used to check options and return False when they prevent the successful start of the source.
options: This optional argument contains the contents of the options() parameter of the syslog-ng OSE configuration object as a Python dictionary.
open(self) method (optional)
The open(self) method opens the resources required for the destination, for example, it initiates a connection to the target service. It is called after init() when syslog-ng OSE is started or reloaded. If send() returns with an error, syslog-ng OSE calls close() and open() before trying to send again.
If open() fails, it should return the False value. In this case, syslog-ng OSE retries it every time-reopen() seconds. By default, this is 1 second for Python sources and destinations, the value of time-reopen() is not inherited from the global option. For details, see Error handling in the python() destination.
send(self, message) method (mandatory)
The send method sends a message to the target service. It should return True to indicate success, or self.QUEUED when using batch mode. For other possible return values, see the description of the flush() method. Note that for batch mode, the flush() method must be implemented as well.
This is the only mandatory method of the destination.
If a message cannot be delivered after the number of times set in retries() (by default: 3), syslog-ng OSE drops the message and continues with the next message. For details, see Error handling in the python() destination.
The method can return True, False, or one of the following constants:
-
self.DROP: The message is dropped immediately.
-
self.ERROR: Corresponds to boolean False. The message is put back to the queue, and sending the message is attempted (up to the number of the retries() option). The destination is suspended for time-reopen() seconds.
-
self.SUCCESS: Corresponds to boolean True. The message was sent successfully.
-
self.QUEUED: The send() method should return this value when using batch mode, if it has successfully added the message to the batch. Message acknowledgment of batches is controlled by the flush() method.
-
self.NOT_CONNECTED: The message is put back to the queue, and the destination is suspended. The open() method will be called, and the sending the messages will be continued with the same message/batch.
-
self.RETRY: The message is put back to the queue, and sending the message is attempted (up to the number of the retries() option). If sending the message has failed retries() times, self.NOT_CONNECTED is returned.
flush(self) method (optional)
Send the messages in a batch. You can use this method to implement batch-mode message sending instead of sending messages one-by-one. When using batch mode, the send() method adds the messages to a batch (for example, a list), and the flush() method sends the messages as configured in the batch-bytes(), batch-lines(), or batch-timeout() options.
The method can return True, False, or one of the following constants:
-
self.DROP: The messages cannot be sent and the entire batch is dropped immediately.
-
self.ERROR: Corresponds to boolean False. The message is put back to the queue, and sending the message is attempted (up to the number of the retries() option). The destination is suspended for time-reopen() seconds.
-
self.SUCCESS: Corresponds to boolean True. The message was sent successfully.
-
self.NOT_CONNECTED: The message is put back to the queue, and the destination is suspended. The open() method will be called, and the sending the messages will be continued with the same message/batch.
-
self.RETRY: The message is put back to the queue, and sending the message is attempted (up to the number of the retries() option). If sending the message has failed retries() times, self.NOT_CONNECTED is returned.
close(self) method (optional)
Close the connection to the target service. Usually it is called right before deinit() when stopping or reloading syslog-ng OSE. It is also called when send() fails.
The deinit(self) method (optional)
This method is executed when syslog-ng OSE is stopped or reloaded. This method does not return a value.
|
|
Caution:
If you reload syslog-ng OSE, existing Python objects are destroyed, therefore the context and state information of Python blocks is lost. Log rotation and updating the configuration of syslog-ng OSE typically involves a reload. |
Error handling in the python() destination
The Python destination handles errors as follows.
-
Currently syslog-ng OSE ignores every error from the open method until the first log message arrives to the Python destination. If the fist message has arrived and there was an error in the open method, syslog-ng OSE starts calling the open method every time-reopen() second, until opening the destination succeeds.
-
If the open method returns without error, syslog-ng OSE calls the send method to send the first message.
-
If the send method returns with an error, syslog-ng OSE calls the is_opened method.
-
If the is_opened method returns an error, syslog-ng OSE starts calling the open method every time-reopen() second, until opening the destination succeeds.
-
Otherwise, syslog-ng OSE calls the send method again.
-
If the send method has returned with an error retries() times and the is_opened method has not returned any errors, syslog-ng OSE drops the message and attempts to process the next message.
Example: Write logs into a file
The purpose of this example is only to demonstrate the basics of the Python destination, if you really want to write log messages into text files, use the file destination instead.
The following sample code writes the body of log messages into the /tmp/example.txt file. Only the send() method is implemented, meaning that syslog-ng OSE opens and closes the file for every message.
destination d_python_to_file {
python(
class("TextDestination")
);
};
log {
source(src);
destination(d_python_to_file);
};
python {
class TextDestination(object):
def send(self, msg):
self.outfile = open("/tmp/example.txt", "a")
self.outfile.write("MESSAGE = %s\n" % msg["MESSAGE"])
self.outfile.flush()
self.outfile.close();
return True
};
The following code is similar to the previous example, but it opens and closes the file using the open() and close() methods.
destination d_python_to_file {
python(
class("TextDestination")
);
};
log {
source(src);
destination(d_python_to_file);
};
python {
class TextDestination(object):
def open(self):
try:
self.outfile = open("/tmp/example.txt", "a")
return True
except:
return False
def send(self, msg):
self.outfile.write("MESSAGE = %s\n" % msg["MESSAGE"])
self.outfile.flush()
return True
def close(self):
try:
self.outfile.flush()
self.outfile.close();
return True
except:
return False
};
For a more detailed example about sending log messages to an MQTT (Message Queuing Telemetry Transport) server, see the Writing Python destination in syslog-ng: how to send log messages to MQTT blog post.
Example: Print logs in batch mode
The following is a simple destination that uses the flush() method to print the messages in batch mode.
class MyDestination(object):
def init(self, options):
self.bulk = list()
return True
def send(self, msg):
self.bulk.append(msg["MSG"].decode())
return self.QUEUED
def flush(self):
print("flushing: " + ",".join(self.bulk))
self.bulk = list()
return self.SUCCESS
For the list of available optional parameters, see python() destination options.
The Python destination allows you to write your own destination in Python. The python() destination has the following options. The class() option is mandatory. For details on writing destinations in Python, see python: writing custom Python destinations.
batch-bytes()
| Accepted values: |
number [bytes] |
| Default: |
none |
Description: Sets the maximum size of payload in a batch. If the size of the messages reaches this value, syslog-ng OSE sends the batch to the destination even if the number of messages is less than the value of the batch-lines() option.
Note that if the batch-timeout() option is enabled and the queue becomes empty, syslog-ng OSE flushes the messages only if batch-timeout() expires, or the batch reaches the limit set in batch-bytes().
Available in syslog-ng OSE version 3.19 and later.
This option does not have any effect unless the flush() method is implemented in the destination.
batch-lines()
Description: Specifies how many lines are flushed to a destination in one batch. The syslog-ng OSE application waits for this number of lines to accumulate and sends them off in a single batch. Increasing this number increases throughput as more messages are sent in a single batch, but also increases message latency.
For example, if you set batch-lines() to 100, syslog-ng OSE waits for 100 messages.
If the batch-timeout() option is disabled, the syslog-ng OSE application flushes the messages if it has sent batch-lines() number of messages, or the queue became empty. If you stop or reload syslog-ng OSE or in case of network sources, the connection with the client is closed, syslog-ng OSE automatically sends the unsent messages to the destination.
Note that if the batch-timeout() option is enabled and the queue becomes empty, syslog-ng OSE flushes the messages only if batch-timeout() expires, or the batch reaches the limit set in batch-lines().
For optimal performance, make sure that the syslog-ng OSE source that feeds messages to this destination is configured properly: the value of the log-iw-size() option of the source must be higher than the batch-lines()*workers() of the destination. Otherwise, the size of the batches cannot reach the batch-lines() limit.
This option does not have any effect unless the flush() method is implemented in the destination.
batch-timeout()
| Type: |
time in milliseconds |
| Default: |
-1 (disabled) |
Description: Specifies the time syslog-ng OSE waits for lines to accumulate in the output buffer. The syslog-ng OSE application sends batches to the destinations evenly. The timer starts when the first message arrives to the buffer, so if only few messages arrive, syslog-ng OSE sends messages to the destination at most once every batch-timeout() milliseconds.
This option does not have any effect unless the flush() method is implemented in the destination.
class()
| Type: |
string |
| Default: |
N/A |
Description: The name of the Python class that implements the destination, for example:
python(
class("MyPythonDestination")
);
If you want to store the Python code in an external Python file, the class() option must include the name of the Python file containing the class, without the path and the .py extension, for example:
python(
class("MyPythonfilename.MyPythonDestination")
);
For details, see Python code in external files
disk-buffer()
Description: This option enables putting outgoing messages into the disk buffer of the destination to avoid message loss in case of a system failure on the destination side. It has the following options:
| reliable() |
| Type: |
yes|no |
| Default: |
no |
|
Description: If set to yes, syslog-ng OSE cannot lose logs in case of reload/restart, unreachable destination or syslog-ng OSE crash. This solution provides a slower, but reliable disk-buffer option. It is created and initialized at startup and gradually grows as new messages arrive. If set to no, the normal disk-buffer will be used. This provides a faster, but less reliable disk-buffer option.
|
|
Caution:
Hazard of data loss! If you change the value of reliable() option when there are messages in the disk-buffer, the messages stored in the disk-buffer will be lost. | |
| compaction() |
| Type: |
yes|no |
| Default: |
no |
|
Description: If set to yes, syslog-ng OSE prunes the unused space in the LogMessage representation, making the disk queue size smaller at the cost of some CPU time. Setting the compaction() argument to yes is recommended when numerous name-value pairs are unset during processing, or when the same names are set multiple times. |
NOTE: Simply unsetting these name-value pairs by using the unset() rewrite operation is not enough, as due to performance reasons that help when syslog-ng is CPU bound, the internal representation of a LogMessage will not release the memory associated with these name-value pairs. In some cases, however, the size of this overhead becomes significant (the raw message size can grow up to four times its original size), which unnecessarily increases the disk queue file size. For these cases, the compaction will drop "unset" values, making the LogMessage representation smaller at the cost of some CPU time required to perform compaction.
| dir() |
| Type: |
string |
| Default: |
N/A |
|
Description: Defines the folder where the disk-buffer files are stored.
|
|
Caution:
When creating a new dir() option for a disk buffer, or modifying an existing one, make sure you delete the persist file.
syslog-ng OSE creates disk-buffer files based on the path recorded in the persist file. Therefore, if the persist file is not deleted after modifying the dir() option, then following a restart, syslog-ng OSE will look for or create disk-buffer files in their old location. To ensure that syslog-ng OSE uses the new dir() setting, the persist file must not contain any information about the destinations which the disk-buffer file in question belongs to. | |
| disk-buf-size() |
| Type: |
number (bytes) |
| Default: |
|
| Description: This is a required option. The maximum size of the disk-buffer in bytes. The minimum value is 1048576 bytes. If you set a smaller value, the minimum value will be used automatically. It replaces the old log-disk-fifo-size() option. |
| mem-buf-length() |
| Type: |
number (messages) |
| Default: |
10000 |
| Description: Use this option if the option reliable() is set to no. This option contains the number of messages stored in overflow queue. It replaces the old log-fifo-size() option. It inherits the value of the global log-fifo-size() option if provided. If it is not provided, the default value is 10000 messages. Note that this option will be ignored if the option reliable() is set to yes. |
| mem-buf-size() |
| Type: |
number (bytes) |
| Default: |
163840000 |
| Description: Use this option if the option reliable() is set to yes. This option contains the size of the messages in bytes that is used in the memory part of the disk buffer. It replaces the old log-fifo-size() option. It does not inherit the value of the global log-fifo-size() option, even if it is provided. Note that this option will be ignored if the option reliable() is set to no. |
| qout-size() |
| Type: |
number (messages) |
| Default: |
64 |
| Description: The number of messages stored in the output buffer of the destination. Note that if you change the value of this option and the disk-buffer already exists, the change will take effect when the disk-buffer becomes empty. |
Options reliable() and disk-buf-size() are required options.
Example: Examples for using disk-buffer()
In the following case reliable disk-buffer() is used.
destination d_demo {
network(
"127.0.0.1"
port(3333)
disk-buffer(
mem-buf-size(10000)
disk-buf-size(2000000)
reliable(yes)
dir("/tmp/disk-buffer")
)
);
};
In the following case normal disk-buffer() is used.
destination d_demo {
network(
"127.0.0.1"
port(3333)
disk-buffer(
mem-buf-length(10000)
disk-buf-size(2000000)
reliable(no)
dir("/tmp/disk-buffer")
)
);
};
frac-digits()
Description: The syslog-ng application can store fractions of a second in the timestamps according to the ISO8601 format. The frac-digits() parameter specifies the number of digits stored. The digits storing the fractions are padded by zeros if the original timestamp of the message specifies only seconds. Fractions can always be stored for the time the message was received.
NOTE: The syslog-ng OSE application can add the fractions to non-ISO8601 timestamps as well.
NOTE: As syslog-ng OSE is precise up to the microsecond, when the frac-digits() option is set to a value higher than 6, syslog-ng OSE will truncate the fraction seconds in the timestamps after 6 digits.
loaders()
| Type: |
list of python modules |
| Default: |
empty list |
Description: The syslog-ng OSE application imports Python modules specified in this option, before importing the code of the Python class. This option has effect only when the Python class is provided in an external Python file. This option has no effect when the Python class is provided within the syslog-ng OSE configuration file (in a python{} block). You can use the loaders() option to modify the import mechanism that imports Python class. For example, that way you can use hy in your Python class.
python(class(usermodule.HyParser) loaders(hy))
log-fifo-size()
| Type: |
number |
| Default: |
Use global setting. |
Description: The number of messages that the output queue can store.
on-error()
| Accepted values: |
drop-message|drop-property|fallback-to-string|
silently-drop-message|silently-drop-property|silently-fallback-to-string |
| Default: |
Use the global setting (which defaults to drop-message) |
Description: Controls what happens when type-casting fails and syslog-ng OSE cannot convert some data to the specified type. By default, syslog-ng OSE drops the entire message and logs the error. Currently the value-pairs() option uses the settings of on-error().
-
drop-message: Drop the entire message and log an error message to the internal() source. This is the default behavior of syslog-ng OSE.
-
drop-property: Omit the affected property (macro, template, or message-field) from the log message and log an error message to the internal() source.
-
fallback-to-string: Convert the property to string and log an error message to the internal() source.
-
silently-drop-message: Drop the entire message silently, without logging the error.
-
silently-drop-property: Omit the affected property (macro, template, or message-field) silently, without logging the error.
-
silently-fallback-to-string: Convert the property to string silently, without logging the error.
options()
| Type: |
string |
| Default: |
N/A |
Description: This option allows you to pass custom values from the configuration file to the Python code. Enclose both the option names and their values in double-quotes. The Python code will receive these values during initialization as the options dictionary. For example, you can use this to set the IP address of the server from the configuration file, so it is not hard-coded in the Python object.
python(
class("MyPythonClass")
options(
"host" "127.0.0.1"
"port" "1883"
"otheroption" "value")
);
For example, you can refer to the value of the host field in the Python code as options["host"]. Note that the Python code receives the values as strings, so you might have to cast them to the type required, for example: int(options["port"])
NOTE: From version 3.27, syslog-ng OSE supports the arrow syntax for declaring custom Java and Python options. You can alternatively declare them using a similar syntax:
options(
"host" => "localhost"
"port" => "1883"
"otheroption" => "value"
)
persist-name()
Description:If you receive the following error message during syslog-ng OSE startup, set the persist-name() option of the duplicate drivers:
Error checking the uniqueness of the persist names, please override it with persist-name option. Shutting down.
This error happens if you use identical drivers in multiple sources, for example, if you configure two file sources to read from the same file. In this case, set the persist-name() of the drivers to a custom string, for example, persist-name("example-persist-name1").
NOTE: Starting with 3.26, syslog-ng OSE assigns a persist name to Python sources and destinations. The persist name is generated from the class name. If you want to use the same Python class multiple times in your syslog-ng OSE configuration, add a unique persist-name() to each source or destination, otherwise syslog-ng OSE will not start. For example:
log {
source { python(class(PyNetworkSource) options("port" "8080") persist-name("<unique-string>); };
source { python(class(PyNetworkSource) options("port" "8081")); };
};
Alternatively, you can include the following line in the Python package: @staticmethod generate_persist_name. For example:
from syslogng import LogSource
class PyNetworSource(LogSource):
@staticmethod
def generate_persist_name(options):
return options["port"]
def run(self):
pass
def request_exit(self):
pass
throttle()
Description: Sets the maximum number of messages sent to the destination per second. Use this output-rate-limiting functionality only when using disk-buffer as well to avoid the risk of losing messages. Specifying 0 or a lower value sets the output limit to unlimited.
value-pairs()
| Type: |
parameter list of the value-pairs() option |
| Default: |
scope("selected-macros" "nv-pairs") |
Description: The value-pairs() option creates structured name-value pairs from the data and metadata of the log message. For details on using value-pairs(), see Structuring macros, metadata, and other value-pairs.
NOTE: Empty keys are not logged.
You can use this option to limit which name-value pairs are passed to the Python code for each message. Note that if you use the value-pairs() option, the Python code receives the specified value-pairs as a Python dict. Otherwise, it receives the message object. In the following example, only the text of the log message is passed to Python.
destination d_python_to_file {
python(
class("pythonexample.TextDestination")
value-pairs(key(MESSAGE))
);
};
The redis() driver sends messages as name-value pairs to a Redis key-value store.
For the list of available parameters, see redis() destination options.
Declaration:
redis(
host("<redis-server-address>")
port("<redis-server-port>")
auth("<redis-server-password>") # Optional, for password-protected servers
command("<redis-command>", "<first-command-parameter>", "<second-command-parameter>", "<third-command-parameter>")
);
Example: Using the redis() driver
The following destination counts the number of log messages received per host.
destination d_redis {
redis(
host("localhost")
port(6379)
command("HINCRBY", "hosts", "$HOST", "1")
);
};
The following example creates a statistic from Apache webserver logs about the browsers that the visitors use (per minute)
@version: 3.33
source s_apache {
file("/var/log/apache2/access.log");
};
parser p_apache {
csv-parser(columns("APACHE.CLIENT_IP", "APACHE.IDENT_NAME", "APACHE.USER_NAME",
"APACHE.TIMESTAMP", "APACHE.REQUEST_URL", "APACHE.REQUEST_STATUS",
"APACHE.CONTENT_LENGTH", "APACHE.REFERER", "APACHE.USER_AGENT",
"APACHE.PROCESS_TIME", "APACHE.SERVER_NAME")
flags(escape-double-char,strip-whitespace)
delimiters(" ")
quote-pairs('""[]')
);
};
destination d_redis {
redis( command("HINCRBY" "${MONTH_ABBREV} ${DAY} ${HOUR}:${MIN}" "${APACHE.USER_AGENT}" "1"));
};
log {
source(s_apache);
parser(p_apache);
destination(d_redis);
};
The redis() driver sends messages as name-value pairs to a Redis key-value store.
The redis() destination has the following options:
auth()
| Type: |
hostname or IP address |
| Default: |
N/A |
Description: The password used for authentication on a password-protected Redis server. Available in syslog-ng OSE version 3.10 and later.
command()
| Type: |
comma-separated list of strings ("<redis-command>", "<first-command-parameter>", "<second-command-parameter>", "<third-command-parameter>") |
| Default: |
empty string |
Description: The Redis command to execute, for example, LPUSH, INCR, or HINCRBY. Using the HINCRBY command with an increment value of 1 allows you to create various statistics. For example, the command("HINCRBY" "${HOST}/programs" "${PROGRAM}" "1") command counts the number of log messages on each host for each program.
Note the following points when using the redis() destination:
-
You can use macros and templates in the parameters of the Redis command.
-
Currently you can use only one command in a redis() destination.
-
The syslog-ng OSE application ignores the return value of the command. If the Redis server returns an error, syslog-ng OSE closes the connection.
disk-buffer()
Description: This option enables putting outgoing messages into the disk buffer of the destination to avoid message loss in case of a system failure on the destination side. It has the following options:
| reliable() |
| Type: |
yes|no |
| Default: |
no |
|
Description: If set to yes, syslog-ng OSE cannot lose logs in case of reload/restart, unreachable destination or syslog-ng OSE crash. This solution provides a slower, but reliable disk-buffer option. It is created and initialized at startup and gradually grows as new messages arrive. If set to no, the normal disk-buffer will be used. This provides a faster, but less reliable disk-buffer option.
|
|
Caution:
Hazard of data loss! If you change the value of reliable() option when there are messages in the disk-buffer, the messages stored in the disk-buffer will be lost. | |
| compaction() |
| Type: |
yes|no |
| Default: |
no |
|
Description: If set to yes, syslog-ng OSE prunes the unused space in the LogMessage representation, making the disk queue size smaller at the cost of some CPU time. Setting the compaction() argument to yes is recommended when numerous name-value pairs are unset during processing, or when the same names are set multiple times. |
NOTE: Simply unsetting these name-value pairs by using the unset() rewrite operation is not enough, as due to performance reasons that help when syslog-ng is CPU bound, the internal representation of a LogMessage will not release the memory associated with these name-value pairs. In some cases, however, the size of this overhead becomes significant (the raw message size can grow up to four times its original size), which unnecessarily increases the disk queue file size. For these cases, the compaction will drop "unset" values, making the LogMessage representation smaller at the cost of some CPU time required to perform compaction.
| dir() |
| Type: |
string |
| Default: |
N/A |
|
Description: Defines the folder where the disk-buffer files are stored.
|
|
Caution:
When creating a new dir() option for a disk buffer, or modifying an existing one, make sure you delete the persist file.
syslog-ng OSE creates disk-buffer files based on the path recorded in the persist file. Therefore, if the persist file is not deleted after modifying the dir() option, then following a restart, syslog-ng OSE will look for or create disk-buffer files in their old location. To ensure that syslog-ng OSE uses the new dir() setting, the persist file must not contain any information about the destinations which the disk-buffer file in question belongs to. | |
| disk-buf-size() |
| Type: |
number (bytes) |
| Default: |
|
| Description: This is a required option. The maximum size of the disk-buffer in bytes. The minimum value is 1048576 bytes. If you set a smaller value, the minimum value will be used automatically. It replaces the old log-disk-fifo-size() option. |
| mem-buf-length() |
| Type: |
number (messages) |
| Default: |
10000 |
| Description: Use this option if the option reliable() is set to no. This option contains the number of messages stored in overflow queue. It replaces the old log-fifo-size() option. It inherits the value of the global log-fifo-size() option if provided. If it is not provided, the default value is 10000 messages. Note that this option will be ignored if the option reliable() is set to yes. |
| mem-buf-size() |
| Type: |
number (bytes) |
| Default: |
163840000 |
| Description: Use this option if the option reliable() is set to yes. This option contains the size of the messages in bytes that is used in the memory part of the disk buffer. It replaces the old log-fifo-size() option. It does not inherit the value of the global log-fifo-size() option, even if it is provided. Note that this option will be ignored if the option reliable() is set to no. |
| qout-size() |
| Type: |
number (messages) |
| Default: |
64 |
| Description: The number of messages stored in the output buffer of the destination. Note that if you change the value of this option and the disk-buffer already exists, the change will take effect when the disk-buffer becomes empty. |
Options reliable() and disk-buf-size() are required options.
Example: Examples for using disk-buffer()
In the following case reliable disk-buffer() is used.
destination d_demo {
network(
"127.0.0.1"
port(3333)
disk-buffer(
mem-buf-size(10000)
disk-buf-size(2000000)
reliable(yes)
dir("/tmp/disk-buffer")
)
);
};
In the following case normal disk-buffer() is used.
destination d_demo {
network(
"127.0.0.1"
port(3333)
disk-buffer(
mem-buf-length(10000)
disk-buf-size(2000000)
reliable(no)
dir("/tmp/disk-buffer")
)
);
};
batch-bytes()
| Accepted values: |
number [bytes] |
| Default: |
none |
Description: Sets the maximum size of payload in a batch. If the size of the messages reaches this value, syslog-ng OSE sends the batch to the destination even if the number of messages is less than the value of the batch-lines() option.
Note that if the batch-timeout() option is enabled and the queue becomes empty, syslog-ng OSE flushes the messages only if batch-timeout() expires, or the batch reaches the limit set in batch-bytes().
Available in syslog-ng OSE version 3.19 and later.
batch-lines()
Description: Specifies how many lines are flushed to a destination in one batch. The syslog-ng OSE application waits for this number of lines to accumulate and sends them off in a single batch. Increasing this number increases throughput as more messages are sent in a single batch, but also increases message latency.
For example, if you set batch-lines() to 100, syslog-ng OSE waits for 100 messages.
If the batch-timeout() option is disabled, the syslog-ng OSE application flushes the messages if it has sent batch-lines() number of messages, or the queue became empty. If you stop or reload syslog-ng OSE or in case of network sources, the connection with the client is closed, syslog-ng OSE automatically sends the unsent messages to the destination.
Note that if the batch-timeout() option is enabled and the queue becomes empty, syslog-ng OSE flushes the messages only if batch-timeout() expires, or the batch reaches the limit set in batch-lines().
For optimal performance, make sure that the syslog-ng OSE source that feeds messages to this destination is configured properly: the value of the log-iw-size() option of the source must be higher than the batch-lines()*workers() of the destination. Otherwise, the size of the batches cannot reach the batch-lines() limit.
batch-timeout()
| Type: |
time in milliseconds |
| Default: |
-1 (disabled) |
Description: Specifies the time syslog-ng OSE waits for lines to accumulate in the output buffer. The syslog-ng OSE application sends batches to the destinations evenly. The timer starts when the first message arrives to the buffer, so if only few messages arrive, syslog-ng OSE sends messages to the destination at most once every batch-timeout() milliseconds.
hook-commands()
Description: This option makes it possible to execute external programs when the relevant driver is initialized or torn down. The hook-commands() can be used with all source and destination drivers with the exception of the usertty() and internal() drivers.
NOTE: The syslog-ng OSE application must be able to start and restart the external program, and have the necessary permissions to do so. For example, if your host is running AppArmor or SELinux, you might have to modify your AppArmor or SELinux configuration to enable syslog-ng OSE to execute external applications.
Using the hook-commands() when syslog-ng OSE starts or stops
To execute an external program when syslog-ng OSE starts or stops, use the following options:
| startup() |
| Type: |
string |
| Default: |
N/A |
|
Description: Defines the external program that is executed as syslog-ng OSE starts. |
| shutdown() |
| Type: |
string |
| Default: |
N/A |
|
Description: Defines the external program that is executed as syslog-ng OSE stops. |
Using the hook-commands() when syslog-ng OSE reloads
To execute an external program when the syslog-ng OSE configuration is initiated or torn down, for example, on startup/shutdown or during a syslog-ng OSE reload, use the following options:
| setup() |
| Type: |
string |
| Default: |
N/A |
|
Description: Defines an external program that is executed when the syslog-ng OSE configuration is initiated, for example, on startup or during a syslog-ng OSE reload. |
| teardown() |
| Type: |
string |
| Default: |
N/A |
|
Description: Defines an external program that is executed when the syslog-ng OSE configuration is stopped or torn down, for example, on shutdown or during a syslog-ng OSE reload. |
Example: Using the hook-commands() with a network source
In the following example, the hook-commands() is used with the network() driver and it opens an iptables port automatically as syslog-ng OSE is started/stopped.
The assumption in this example is that the LOGCHAIN chain is part of a larger ruleset that routes traffic to it. Whenever the syslog-ng OSE created rule is there, packets can flow, otherwise the port is closed.
source {
network(transport(udp)
hook-commands(
startup("iptables -I LOGCHAIN 1 -p udp --dport 514 -j ACCEPT")
shutdown("iptables -D LOGCHAIN 1")
)
);
};
host()
| Type: |
hostname or IP address |
| Default: |
127.0.0.1 |
Description: The hostname or IP address of the Redis server.
port()
| Type: |
number |
| Default: |
6379 |
Description: The port number of the Redis server.
retries()
| Type: |
number (of attempts) |
| Default: |
3 |
Description: The number of times syslog-ng OSE attempts to send a message to this destination. If syslog-ng OSE could not send a message, it will try again until the number of attempts reaches retries, then drops the message.
throttle()
Description: Sets the maximum number of messages sent to the destination per second. Use this output-rate-limiting functionality only when using disk-buffer as well to avoid the risk of losing messages. Specifying 0 or a lower value sets the output limit to unlimited.
