tencent cloud

masukan

Separator Format

Terakhir diperbarui:2024-01-20 17:14:28

    Overview

    In a separator log, the entire log data can be structured according to the specified separator, and each complete log ends with a line break \\n. When CLS processes separator logs, you need to define a unique key for each separate field.

    Prerequisites

    Suppose your raw log data is:
    10.20.20.10 - ::: [Tue Jan 22 14:49:45 CST 2019 +0800] ::: GET /online/sample HTTP/1.1 ::: 127.0.0.1 ::: 200 ::: 647 ::: 35 ::: http://127.0.0.1/
    If the separator for log parsing is specified as :::, the log will be segmented into eight fields, and a unique key will be defined for each of them.
    IP: 10.20.20.10 -
    bytes: 35
    host: 127.0.0.1
    length: 647
    referer: http://127.0.0.1/
    request: GET /online/sample HTTP/1.1
    status: 200
    time: [Tue Jan 22 14:49:45 CST 2019 +0800]

    Directions

    Logging in to the console

    1. Log in to the CLS console.
    2. On the left sidebar, click Log Topic to go to the log topic management page.

    Creating a log topic

    1. Click Create Log Topic.
    2. In the pop-up dialog box, enter test-separator as Log Topic Name and click Confirm.

    Managing the machine group

    1. After the log topic is created successfully, click its name to go to the log topic management page.
    2. Click the Collection Configuration tab, click Add in LogListener Collection Configuration, and select the format in which you need to collect logs.
    3. On the Machine Group Management page, select the server group to which to bind the current log topic and click Next to proceed to collection configuration. For more information, see Machine Group Management.

    Configuring collection

    On the Collection Configuration page, enter the collection rule name and enter the Collection Path according to the log collection path format. Log collection path format: [directory prefix expression]/**/[filename expression].
    After the log collection path is entered, LogListener will match all common prefix paths that meet the [directory prefix expression] rule and listen for all log files in the directories (including subdirectories) that meet the [filename expression] rule. The parameters are as detailed below:
    Parameter
    Description
    Directory Prefix
    Directory prefix for log files, which supports only the wildcard characters * and ?.
    \\* indicates to match any multiple characters.
    ? indicates to match any single character.
    /**/
    Current directory and all its subdirectories.
    File Name
    Log file name, which supports only the wildcard characters * and ?.
    \\* indicates to match any multiple characters.
    ? indicates to match any single character.
    Common configuration modes are as follows:
    [Common directory prefix]/**/[common filename prefix]*
    [Common directory prefix]/**/*[common filename suffix]
    [Common directory prefix]/**/[common filename prefix]*[common filename suffix]
    [Common directory prefix]/**/*[common string]*
    Below are examples:
    No.
    Directory Prefix Expression
    Filename Expression
    Description
    1.
    /var/log/nginx
    access.log
    In this example, the log path is configured as /var/log/nginx/**/access.log. LogListener will listen for log files named access.log in all subdirectories in the /var/log/nginx prefix path.
    2.
    /var/log/nginx
    *.log
    In this example, the log path is configured as /var/log/nginx/**/*.log. LogListener will listen for log files suffixed with .log in all subdirectories in the /var/log/nginx prefix path.
    3.
    /var/log/nginx
    error*
    In this example, the log path is configured as /var/log/nginx/**/error*. LogListener will listen for log files prefixed with error in all subdirectories in the /var/log/nginx prefix path.
    Note:
    Only LogListener 2.3.9 and later support adding multiple collection paths.
    The system does not support uploading logs with contents in multiple text formats, which may cause write failures, such as key:"{"substream":XXX}".
    We recommend you configure the collection path as log/*.log and rename the old file after log rotation as log/*.log.xxxx.
    By default, a log file can only be collected by one log topic. If you want to have multiple collection configurations for the same file, add a soft link to the source file and add it to another collection configuration.

    Configuring the collection policy

    Full collection: When LogListener collects a file, it starts reading data from the beginning of the file.
    Incremental collection: When LogListener collects a file, it collects only the newly added content in the file.

    Configuring the separator mode

    1. Set Extraction Mode to Separator.
    2. Select Separator, enter a sample log in the Log Sample text box, and click Extract. The system segments the sample log according to the selected separator and displays it in the extraction result box. You need to define a unique key for each field. Currently, log collection supports a variety of separators. Common separators include space, tab, comma, semicolon, and vertical bar. If your log data uses other separators such as :::, it can also be parsed through custom delimiter.

    Configuring the collection time

    Note:
    The log time is measured in seconds. If the log time is entered in an incorrect format, the collection time is used as the log time.
    The time attribute of a log is defined in two ways: collection time and original timestamp.
    Collection time: The time attribute of a log is determined by the time when CLS collects the log.
    Original timestamp: The time attribute of a log is determined by the timestamp in the raw log.
    Using the collection time as the time attribute of logs Keep Collection Time as enabled.
    Using the original timestamp as the time attribute of logs Disable Collection Time and enter the time key of the original timestamp and the corresponding time parsing format in Time Key and Time Parsing Format respectively. For more information on the time parsing format, see Configuring Time Format.
    Below are examples of how to enter a time parsing format: Example 1: The parsing format of the original timestamp 10/Dec/2017:08:00:00 is %d/%b/%Y:%H:%M:%S. Example 2: The parsing format of the original timestamp `2017-12-10 08:00:00` is %Y-%m-%d %H:%M:%S. Example 3: The parsing format of the original timestamp 12/10/2017, 08:00:00 is %m/%d/%Y, %H:%M:%S.
    Note:
    Second can be used as the unit of log time. If the time is entered in a wrong format, the collection time is used as the log time.

    Configuring filter rules

    Filters are designed to help you extract valuable log data by adding log collection filter rules based on your business needs. If the filter rule is a Perl regular expression, the created filter rule will be used for matching; in other words, only logs that match the regular expression will be collected and reported.
    For separator-formatted logs, you need to configure a filter rule according to the defined custom key-value pair. For example, if you want to collect all log data with a status field with the value 400 or 500 after the sample log is parsed in separator mode, you need to configure key as status and the filter rule as 400|500.
    Note:
    The relationship logic between multiple filter rules is "AND". If multiple filter rules are configured for the same key name, previous rules will be overwritten.

    Configuring parsing-failed log upload

    We recommend you enable Upload Parsing-Failed Logs. After it is enabled, LogListener will upload all types of parsing-failed logs. If it is disabled, such logs will be discarded.
    After this feature is enabled, you need to configure the Key value for parsing failures (which is LogParseFailure by default). All parsing-failed logs are uploaded with the input content as the key name (Key) and the raw log content as the key value (Value).

    Configuring indexes

    1. Click Next to enter the Index Configuration page.
    2. On the Index Configuration page, set the following information:
    Index Status: Select whether to enable it.
    Note:
    Index configuration must be enabled before you can perform searches.
    Full-Text Index: Select whether to set it to case-sensitive.
    Full-Text Delimiter: The default value is @&()='",;:<>[]{}/ \\n\\t\\r and can be modified as needed.
    Key-Value Index: Disabled by default. You can configure the field type, delimiters, and whether to enable statistical analysis according to the key name as needed. To enable key-value index, you can set
    
    to
    
    .
    3. Click Submit.
    For more information on log search, see Overview and Syntax Rules.
    Hubungi Kami

    Hubungi tim penjualan atau penasihat bisnis kami untuk membantu bisnis Anda.

    Dukungan Teknis

    Buka tiket jika Anda mencari bantuan lebih lanjut. Tiket kami tersedia 7x24.

    Dukungan Telepon 7x24