Chat now with support
Chat with Support

syslog-ng Premium Edition 7.0.34 - Administration Guide

Preface Introduction to syslog-ng The concepts of syslog-ng Installing syslog-ng PE The syslog-ng PE quick-start guide The syslog-ng PE configuration file Collecting log messages — sources and source drivers
How sources work default-network-drivers: Receive and parse common syslog messages internal: Collecting internal messages file: Collecting messages from text files google-pubsub: collecting messages from the Google Pub/Sub messaging service wildcard-file: Collecting messages from multiple text files linux-audit: Collecting messages from Linux audit logs mssql, oracle, sql: collecting messages from an SQL database network: Collecting messages using the RFC3164 protocol (network() driver) office365: Fetching logs from Office 365 osquery: Collect and parse osquery result logs pipe: Collecting messages from named pipes program: Receiving messages from external applications python: writing server-style Python sources python-fetcher: writing fetcher-style Python sources snmptrap: Read Net-SNMP traps syslog: Collecting messages using the IETF syslog protocol (syslog() driver) system: Collecting the system-specific log messages of a platform systemd-journal: Collecting messages from the systemd-journal system log storage systemd-syslog: Collecting systemd messages using a socket tcp, tcp6,udp, udp6: Collecting messages from remote hosts using the BSD syslog protocol udp-balancer: Receiving UDP messages at very high rate unix-stream, unix-dgram: Collecting messages from UNIX domain sockets windowsevent: Collecting Windows event logs
Sending and storing log messages — destinations and destination drivers
elasticsearch2>: Sending messages directly to Elasticsearch version 2.0 or higher (DEPRECATED) elasticsearch-http: Sending messages to Elasticsearch HTTP Event Collector file: Storing messages in plain-text files google_bigquery(): Sending logs to a Google BigQuery table google_bigquery_managedaccount(): Sending logs to a Google BigQuery table authenticated by Google Cloud managed service account google_pubsub(): Sending logs to the Google Cloud Pub/Sub messaging service google_pubsub-managedaccount(): Sending logs to the Google Cloud Pub/Sub messaging service authenticated by Google Cloud managed service account hdfs: Storing messages on the Hadoop Distributed File System (HDFS) http: Posting messages over HTTP kafka(): Publishing messages to Apache Kafka (Java implementation) (DEPRECATED) kafka-c(): Publishing messages to Apache Kafka using the librdkafka client (C implementation) logstore: Storing messages in encrypted files mongodb: Storing messages in a MongoDB database network: Sending messages to a remote log server using the RFC3164 protocol (network() driver) pipe: Sending messages to named pipes program: Sending messages to external applications python: writing custom Python destinations sentinel(): Sending logs to the Microsoft Azure Sentinel cloud snmp: Sending SNMP traps smtp: Generating SMTP messages (email) from logs splunk-hec: Sending messages to Splunk HTTP Event Collector sql(): Storing messages in an SQL database stackdriver: Sending logs to the Google Stackdriver cloud syslog: Sending messages to a remote logserver using the IETF-syslog protocol syslog-ng(): Forward logs to another syslog-ng node tcp, tcp6, udp, udp6: Sending messages to a remote log server using the legacy BSD-syslog protocol (tcp(), udp() drivers) unix-stream, unix-dgram: Sending messages to UNIX domain sockets usertty: Sending messages to a user terminal — usertty() destination Client-side failover
Routing messages: log paths, flags, and filters Global options of syslog-ng PE TLS-encrypted message transfer Advanced Log Transport Protocol Reliability and minimizing the loss of log messages Manipulating messages parser: Parse and segment structured messages Processing message content with a pattern database Correlating log messages Enriching log messages with external data Monitoring statistics and metrics of syslog-ng Multithreading and scaling in syslog-ng PE Troubleshooting syslog-ng Best practices and examples The syslog-ng manual pages Glossary

Options of the kafka-c() destination

With the kafka-c() destination of syslog-ng PE, you can directly publish log messages to the Apache Kafka message bus, where subscribers can access them. The kafka-c() destination has the following options.

Required options

The following options are required:

  • bootstrap-servers()

  • topic().

batch-lines()
Type: number [lines]
Default: 1

Description: Specifies how many lines are flushed to a destination in one batch. The syslog-ng PE 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 PE waits for 100 messages.

If the batch-timeout() option is disabled, the syslog-ng PE 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 PE or in case of network sources, the connection with the client is closed, syslog-ng PE automatically sends the unsent messages to the destination.

If the batch-timeout() option is enabled and the queue becomes empty, syslog-ng PE 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 PE 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.

NOTE: The syslog-ng PE configuration accepts this option with sync-send() set to both "yes" or "no", but the option will only take effect if you set sync-send() to "yes".

NOTE: If you set sync-send() to "yes", the number you specify for batch-lines() affects how many messages syslog-ng PE packs into once transaction.

batch-timeout()
Type: time [milliseconds]
Default: -1 (disabled)

Description: Specifies the time syslog-ng PE waits for lines to accumulate in the output buffer. The syslog-ng PE 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 PE sends messages to the destination once every batch-timeout() milliseconds at most.

NOTE: The syslog-ng PE configuration accepts this option with sync-send() set to both "yes" or "no", but the option will only take effect if you set sync-send() to "yes".

NOTE: When setting batch-timeout(), consider the value of the transaction.timeout.ms Kafka property. If in case of timeout (that is, if syslog-ng PE does not receive batch-lines() amount of messages) the value of batch-timeout() exceeds the value of transaction.timeout.ms, syslog-ng PE will not send out messages in time.

For more information about the default values of the transaction.timeout.ms Kafka property, see the librdkafka documentation.

bootstrap-servers()
Type: string
Default: N/A

Description: Required option. Specifies the hostname or IP address of the Kafka server. When specifying an IP address, IPv4 (for example, 192.168.0.1) or IPv6 (for example, [::1]) can be used as well. Use a colon (:) after the address to specify the port number of the server. When specifying multiple addresses, use a comma to separate the addresses, for example, bootstrap-servers("127.0.0.1:2525,remote-server-hostname:6464")

config()
Type: N/A
Default: N/A

Description: Advanced configuration option to fine-tune all properties of the official Kafka producer. For details, see the librdkafka documentation.

The syntax of the config() option is the following:

config (
  "acks" => "all"
  "compression.type" => "snappy"
)
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 PE cannot lose logs in case of reload/restart, unreachable destination or syslog-ng PE crash. This solution provides a slower, but more 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 PE 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.

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 PE 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 PE will look for or create disk-buffer files in their old location. To ensure that syslog-ng PE 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: 1000

Description: The number of messages stored in the output buffer of the destination.

NOTE: 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.
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")
        )
    );
};
truncate-size-ratio()
Type: number (between 0 and 1)
Default: 0.1 (10%)

Description: Limits the truncation of the disk-buffer file. Truncating the disk-buffer file can slow down the disk IO operations, but it saves disk space, so syslog-ng only truncates the file, if the possible disk gain is more than truncate-size-ratio() times disk-buf-size().

 Caution:

One Identity does not recommend you to change truncate-size-ratio(). Only change its value if you know the performance implications of doing so.
fallback-topic()
Type: string
Default: N/A

Description: If the resolved topic() template is not a valid Kafka topic , syslog-ng PE will use fallback-topic() to send messages.

NOTE: If instead of strings, you use actual templates (that is, a macro like ${MESSAGE}, or a template function like $(format-json)) in the topic() option, configuring the fallback-topic() option is required.

frac-digits()
Type: number
Default: 0

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 PE application can add the fractions to non-ISO8601 timestamps as well.

NOTE: As syslog-ng PE is precise up to the microsecond, when the frac-digits() option is set to a value higher than 6, syslog-ng PE will truncate the fraction seconds in the timestamps after 6 digits.

flush-timeout-on-reload()
Type: integer in milliseconds
Default: 1000

Description: When syslog-ng PE reloads, the Kafka client will also reload.

The flush-timeout-on-reload() option specifies the number of milliseconds syslog-ng PE waits for the Kafka client to flush out in-flight messages. In-flight messages may be:

  • messages that are passed to the Kafka client for sending, which have been sent, but not delivered

  • messages not yet sent out.

flush-timeout-on-shutdown()
Type: integer in milliseconds
Default: 60000

Description: When syslog-ng PE shuts down, the Kafka client will also shut down.

The flush-timeout-on-shutdown() option specifies the number of milliseconds syslog-ng PE waits for the Kafka client to flush out in-flight messages. In-flight messages may be:

  • Messages passed to the Kafka client for sending, already sent, but not yet delivered.

  • Messages not yet sent by the Kafka client.

NOTE: To avoid losing messages, One Identity recommends that you use the sync-send() option set to "yes" in addition to using the disk-buffer() option.

hook-commands()

Description: This option makes it possible to run 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 PE 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 PE to run external applications.

Using hook-commands() when syslog-ng PE starts or stops

To run an external program when syslog-ng PE starts or stops, use the following options:

startup()

Type:

string

Default:

N/A

Description: Defines the external program that is run as syslog-ng PE starts.

shutdown()

Type:

string

Default:

N/A

Description: Defines the external program that is run as syslog-ng PE stops.
Using hook-commands() when syslog-ng PE reloads

To run an external program when the syslog-ng PE configuration is initiated or torn down, for example, on startup/shutdown or during a syslog-ng PE reload, use the following options:

setup()

Type:

string

Default:

N/A

Description: Defines an external program that is run when the syslog-ng PE configuration is initiated, for example, on startup or during a syslog-ng PE reload.

teardown()

Type:

string

Default:

N/A

Description: Defines an external program that is run when the syslog-ng PE configuration is stopped or torn down, for example, on shutdown or during a syslog-ng PE reload.
Example: Using 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 PE 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 PE 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")
        )
     );
};
key()
Type: template
Default: empty string

Description: The key of the partition under which the message is published. You can use templates to change the topic dynamically based on the source or the content of the message, for example, key("${PROGRAM}").

log-fifo-size()
Type: number
Default: Use global setting.

Description: The number of messages that the output queue can store.

local-time-zone()
Type: name of the timezone, or the timezone offset
Default: The local timezone.

Description: Sets the timezone used when expanding filename and tablename templates.

The timezone can be specified by using the name, for example, time-zone("Europe/Budapest")), or as the timezone offset in +/-HH:MM format, for example, +01:00). On Linux and UNIX platforms, the valid timezone names are listed under the /usr/share/zoneinfo directory.

message()
Type: message template
Default: $ISODATE $HOST $MSGHDR$MSG

Description: The message as published to Apache Kafka. You can use templates and template functions (for example, format-json()) to format the message, for example, message("$(format-json --scope rfc5424 --exclude DATE --key ISODATE)").

For details on formatting messages in JSON format, see format-json.

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 PE cannot convert some data to the specified type. By default, syslog-ng PE 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 PE.

  • 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.

persist-name()
Type: string
Default:

Description:If you receive the following error message during syslog-ng PE 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").

poll-timeout()
Type: integer in milliseconds
Default: 1000

Description: Specifies the frequency your syslog-ng PE queries the Kafka client about the amount of messages sent since the last poll-timeout (). In case of multithreading, the first syslog-ng PE worker is responsible for poll-timeout().

retries()
Type: number (of attempts)
Default: 3

Description: If syslog-ng PE cannot send a message, it will try again until the number of attempts reaches retries().

If the number of attempts reaches retries(), syslog-ng PE will wait for time-reopen() time, then tries sending the message again.

send-time-zone()
Accepted values: name of the timezone, or the timezone offset
Default: local timezone

Description: Specifies the time zone associated with the messages sent by syslog-ng, if not specified otherwise in the message or in the destination driver. For details, see Timezones and daylight saving.

The timezone can be specified by using the name, for example, time-zone("Europe/Budapest")), or as the timezone offset in +/-HH:MM format, for example, +01:00). On Linux and UNIX platforms, the valid timezone names are listed under the /usr/share/zoneinfo directory.

sync-send()
Type: yes | no
Default: no

Description: When sync-send() is set to "yes", syslog-ng PE sends the message reliably: it sends a message to the Kafka server, then waits for a reply. In case of failure, syslog-ng PE repeats sending the message, as set in the retries() parameter. If sending the message fails for retries() times, syslog-ng PE will wait for time-reopen() time, then tries sending the message again.

This method ensures reliable message transfer, but is very slow.

When sync-send() is set to "no", syslog-ng PE sends messages asynchronously, and receives the response asynchronously. In case of a problem, syslog-ng PE cannot resend the messages.

NOTE: The underlying Kafka client (that is, librdkafka) may retry sending messages to syslog-ng PE independently several times.

This method is fast, but the transfer is not reliable. Several thousands of messages can be lost before syslog-ng PE recognizes the error.

 Caution:

Hazard of data loss!

If sync-send() is set to "no", the messages passed to the Kafka client can be lost.

To avoid data loss, One Identity recommends that you set sync-send() to "yes", as this setting delivers messages to the Kafka client more reliably.

NOTE: If you want to use the sync-send() option set to "yes", One Identity recommends that you use a Kafka server with version number 0.11.0 or higher.

throttle()
Type: number
Default: 0

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.

time-reopen()
Type: number (seconds)
Default: 60

Description: Optional parameter.

If message sending fails, syslog-ng PE retries sending the messages for retries() time (3 times by default) before waiting for time-reopen() time to try sending it again.

time-zone()
Type: name of the timezone, or the timezone offset
Default: unspecified

Description: Convert timestamps to the timezone specified by this option. If this option is not set, then the original timezone information in the message is used. Converting the timezone changes the values of all date-related macros derived from the timestamp, for example, HOUR. For the complete list of such macros, see Date-related macros.

The timezone can be specified by using the name, for example, time-zone("Europe/Budapest")), or as the timezone offset in +/-HH:MM format, for example, +01:00). On Linux and UNIX platforms, the valid timezone names are listed under the /usr/share/zoneinfo directory.

topic()
Type: string or template
Default: N/A

Description: Required option. The Kafka topic under which the message is published. You can use templates to change the topic dynamically based on the source or the content of the message, for example, topic("${HOST}").

NOTE: Valid topic names for the topic() and fallback-topic() options have the following limitations:

  • The topic name must contain characters within the pattern [-._a-zA-Z0-9].

  • The length of the topic name must be between 1 and 249 characters.

NOTE: If you use templates with the topic() option, configuring the fallback-topic() option is also required.

ts-format()
Type: rfc3164, bsd, rfc3339, iso
Default: rfc3164

Description: Override the global timestamp format (set in the global ts-format() parameter) for the specific destination. For details, see ts-format().

NOTE: This option applies only to file and file-like destinations. Destinations that use specific protocols (for example, network(), or syslog()) ignore this option. For protocol-like destinations, use a template locally in the destination, or use the proto-template option.

workers()
Type: integer
Default: 1

Description: The workers are only responsible for formatting the messages that need to be delivered to the Kafka clients. Configure this option only if your Kafka clients have many threads and they do not receive enough messages. If you set the sync-send() option to yes, the number of workers is automatically set to 1.

NOTE: Kafka clients have their own threadpool, entirely independent from any syslog-ng PE settings. The workers() option has no effect on this threadpool.

logstore: Storing messages in encrypted files

The syslog-ng PE application can store log messages securely in encrypted, compressed and timestamped binary files. Timestamps can be requested from an external Timestamping Authority (TSA).

Logstore files consist of individual chunks, every chunk can be encrypted, compressed, and timestamped separately. Chunks contain compressed log messages and header information needed for retrieving messages from the logstore file.

The syslog-ng PE application generates an SHA-1 hash for every chunk to verify the integrity of the chunk. The hashes of the chunks are chained together to prevent injecting chunks into the logstore file. The syslog-ng PE application can encrypt the logstore using various algorithms, using the aes128 encryption algorithm in CBC mode and the hmac-sha1 hashing (HMAC) algorithm as default. For other algorithms, see cipher() and digest().

The destination filename may include macros which get expanded when the message is written, thus a simple logstore() driver may create several files. For more information on available macros see Macros of syslog-ng PE.

If the expanded filename refers to a directory which does not exist, it will be created depending on the create-dirs() setting (both global and a per destination option).

The logstore() has a single required parameter that specifies the filename that stores the log messages. For the list of available optional parameters, see logstore() destination options.

 Caution:

Hazard of data loss! If your log files are on an NFS-mounted network file system, see Using syslog-ng PE with NFS or CIFS (or SMB) file system for log files.

Declaration
logstore(filename options());
Example: Using the logstore() driver

A simple example saving and compressing log messages.

destination d_logstore { logstore("/var/log/messages.lgs" compress(5) ); };

A more detailed example that encrypts messages, modifies the parameters for closing chunks, and sets file privileges.

destination d_logstore {
    logstore("/var/log/messages-logstore.lgs"
        encrypt-certificate("/opt/syslog-ng/etc/syslog-ng/keys/10-100-20-40/public-certificate-of-the-server.pem")
        owner("example")
        group("example")
        perm(0777)
    );
};

The URL to the Timestamping Authority and if needed, the OID of the timestamping policy can be set as global options, or also per logstore destination. The following example specifies the URL and the OID as global options:

options {
        timestamp-url("http://10.50.50.50:8080/");
        timestamp-policy("0.4.0.2023.1.1");
};

NOTE: When using the logstore() destination, update the configuration of your log rotation program to rotate these files. Otherwise, the log files can become very large.

 Caution:

Since the state of each created file must be tracked by syslog-ng, it consumes some memory for each file. If no new messages are written to a file within 60 seconds (controlled by the time-reap() global option), it is closed, and its state is freed.

Exploiting this, a DoS attack can be mounted against the system. If the number of possible destination files and its needed memory is more than the amount available on the syslog-ng server.

The most suspicious macro is ${PROGRAM}, where the number of possible variations is rather high. Do not use the ${PROGRAM} macro in insecure environments.

Displaying the contents of logstore files

To display the contents of a logstore file, use the lgstool (formerly called logcat) command supplied with syslog-ng, for example, lgstool cat /var/log/messages.lgs. Log messages available in the journal file of the logstore (but not yet written to the logstore file itself) are displayed as well.

To display the contents of encrypted log files, specify the private key of the certificate used to encrypt the file, for example, lgstool cat -k private.key /var/log/messages.lgs. The contents of the file are sent to the standard output, so it is possible to use grep and other tools to find particular log messages, for example, lgstool cat /var/log/messages.lgs |grep 192.168.1.1. For further details, see The logstore tool manual page.

TIP: The lgstool utility is available for Microsoft Windows operating systems at the [[[Undefined variable General.downloads-text]]].

 Caution:

For files that are in use by syslog-ng, the last chunk that is open cannot be read.

Journal files

The syslog-ng PE application processes log messages into a journal file before writing them to the logstore file. That way logstore files are consistent even if syslog-ng PE crashes unexpectedly, avoiding losing messages. Note that this does not protect against losing messages if the operating system crashes.

A journal file is automatically created for every logstore file that syslog-ng PE opens. A journal file consists of journal blocks that store the log messages. When a journal block fills up with messages, syslog-ng PE writes the entire block into the logstore file and starts to reuse the journal block (one journal block becomes one chunk in the logstore file).

If the messages cannot be written to the logstore file (for example, because the disk becomes inaccessible, or file operations are slow), messages are put to the next journal block (syslog-ng PE uses four blocks by default). When all journal blocks become full, syslog-ng PE will stop processing incoming traffic. syslog-ng PE starts accepting messages to the logstore file again when the first journal block is successfully written to the logstore file. If syslog-ng PE receives a HUP or STOP signal, or no new message arrives into the logstore for the period set in the time-reap() parameter, it writes every journal block to the logstore.

When syslog-ng PE is restarted, it automatically processes the journal files to the logstore files, unless a particular logstore file is not part of configuration of syslog-ng PE. Such orphaned journal files can be processed with the lgstool recover command. For details on processing orphaned journal files, see The logstore tool manual page.

 Caution:

  • If a particular logstore destination receives messages at a constant but very low message rate (for example, a 100-byte message every 30 seconds), messages do not get written to the logstore file for a long time, because the journal block does not get full, and messages are more frequent than the time-reap() time. This becomes a problem when using logrotate to rotate the logstore files, because log messages will not be in the files they are expected. To avoid this situation, either use time-based macros in the filenames of the logstore files, or send a HUP signal to syslog-ng PE right before rotating the logstore files.

  • When every block of a journal becomes full and syslog-ng PE stops processing incoming traffic, it will not read new messages at all until a block is successfully written to the related logstore file. This is in contrast with flow-control, where only messages from the source related to the particular destination are not processed.

  • The messages in the journal file are in plain-text format. They are neither encrypted nor compressed. The journal file has the same permission as the logstore: by default, root privileges are required to access them. Make sure you consider this if you change the permissions of the journal file (owner, group, perm) in the syslog-ng PE configuration file.

NOTE: Journal files are located in the same folder as the logstore file. The name of the journal file is the same as the logstore file with .jor suffix added. For example, the journal file for messages.lgs is messages.lgs.jor.

The syslog-ng PE application uses a separate journal file for every logstore file. Every journal file is processed by a separate thread. The journal files are mapped into the memory. The journal of an individual logstore file uses up to journal-block-size()*journal-block-count() memory address, which is 4MB by default. However, if you have several logstore files open in parallel (for example, you are collecting log messages from 500 hosts and storing them in separate files for every host, and the hosts are continuously sending messages), the memory requirements for journaling rise quickly (to approximately 2GB for the 500 hosts). To limit the memory use of journals, adjust the logstore-journal-shmem-threshold() global option (by default, it is 512MB).

If the memory required for the journal files exceeds the logstore-journal-shmem-threshold() limit, syslog-ng PE will store only a single journal block of every journal file in the memory, and — if more blocks are needed for a journal — store the additional blocks on the hard disk. Opening new logstore files means allocating memory for one new journal block for every new file. In extreme situations involving large traffic, this can lead to syslog-ng PE consuming the entire memory of the system. Adjust the journal-block-size() and your file-naming conventions as needed to avoid such situations. For details on logstore journals, see Journal files.

 Caution:

If you have a large amount of open logstore files in parrallel (for example, you are using the ${HOST} or ${PROGRAM} macros in your filenames) consider lowering the journal-block-size() to avoid syslog-ng PE consuming the entire memory of the system.

Example: Calculating memory usage of logstore journals

If you are using the default settings (4 journal blocks for every logstore journal, one block is 1MB, logstore-journal-shmem-threshold() is 512MB), this means that syslog-ng PE will allocate 4MB memory for every open logstore file, up to 512MB if you have 128 open logstore files. Opening a new logstore file would require 4 more megabytes of memory for journaling, bringing the total required memory to 516MB, which is above the logstore-journal-shmem-threshold(). In this case, syslog-ng PE switches to storing only a single journal block in the memory, lowering the memory requirements of journaling to 129MB. However, opening more and more logstore files will require more and more memory, and this is not limited, except when syslog-ng PE reaches the maximum number of files that can be open (as set in the --fd-limit command-line option).

Related Documents

The document was helpful.

Select Rating

I easily found the information I needed.

Select Rating