Use the syslog protocol to upload logs

Limits

• Syslog logs must conform to the RFC 5424 protocol.
• The maximum size of each log entry is 64 KB.
• Transport Layer Security (TLS) 1.2 must be used for secure data transmission.

Configurations

If you upload logs by using the syslog protocol, you must configure the address to which the logs are uploaded. The address is in the Log Service endpoint:syslog port format, for example, cn-hangzhou-intranet.log.aliyuncs.com:10009. Specify an endpoint based on the region where your Log Service project resides. For more information, see Service endpoint. The syslog port is 10009. In addition, you must specify a Log Service project, Logstore, and AccessKey pair in the STRUCTURED-DATA field. The following table describes the fields.

Example 1: Use the rsyslog utility to upload syslog logs to Log Service

1. Open the rsyslog configuration file.
   The default path of the rsyslog configuration file is /etc/rsyslog.conf.
2. Configure the following fields based on your rsyslog version, and add the configurations to the end of your rsyslog configuration file:
   • Rsyslog v8 or later

     Set $DefaultNetstreamDriverCAFile to the path of the root certificate in the system.

     # Setup disk assisted queues 
     $WorkDirectory /var/spool/rsyslog # where to place spool files 
     $ActionQueueFileName fwdRule1     # unique name prefix for spool files 
     $ActionQueueMaxDiskSpace 1g       # 1gb space limit (use as much as possible) 
     $ActionQueueSaveOnShutdown on     # save messages to disk on shutdown 
     $ActionQueueType LinkedList       # run asynchronously 
     $ActionResumeRetryCount -1        # infinite retries if host is down 
     $ActionSendTCPRebindInterval 100  # close and re-open the connection to the remote host every 100 of messages sent. 
     #RsyslogGnuTLS set to default ca path 
     $DefaultNetstreamDriverCAFile /etc/ssl/certs/ca-bundle.crt 
     template(name="LogServiceFormat" type="string" 
     string="<%pri%>1 %timestamp:::date-rfc3339% %HOSTNAME% %app-name% %procid% %msgid% [logservice project=\"test-project-1\" logstore=\"test-logstore-1\" access-key-id=\"<yourAccessKeyId>\" access-key-secret=\"<yourAccessKeySecret>\"] %msg%\n" 
     ) 
     # Send messages to Loggly over TCP using the template. 
     action(type="omfwd" protocol="tcp" target="cn-hangzhou.log.aliyuncs.com" port="10009" template="LogServiceFormat" StreamDriver="gtls" StreamDriverMode="1" StreamDriverAuthMode="x509/name" StreamDriverPermittedPeers="*.log.aliyuncs.com")

   • Rsyslog v7 or earlier

     Set $DefaultNetstreamDriverCAFile to the path of the root certificate in the system.

     # Setup disk assisted queues 
     $WorkDirectory /var/spool/rsyslog       # where to place spool files 
     $ActionQueueFileName fwdRule1           # unique name prefix for spool files $ActionQueueMaxDiskSpace 1g             # 1gb space limit (use as much as possible) $ActionQueueSaveOnShutdown on           # save messages to disk on shutdown 
     $ActionQueueType LinkedList             # run asynchronously 
     $ActionResumeRetryCount -1              # infinite retries if host is down $ActionSendTCPRebindInterval 100        # close and re-open the connection to the remote host every 100 of messages sent. 
     # RsyslogGnuTLS set to default ca path 
     $DefaultNetstreamDriverCAFile /etc/ssl/certs/ca-bundle.crt 
     $ActionSendStreamDriver gtls 
     $ActionSendStreamDriverMode 1 
     $ActionSendStreamDriverAuthMode x509/name 
     $ActionSendStreamDriverPermittedPeer cn-hangzhou.log.aliyuncs.com 
     template(name="LogServiceFormat" type="string" string="<%pri%>1 %timestamp:::date-rfc3339% %HOSTNAME% %app-name% %procid% %msgid% [logservice project=\"test-project-1\" logstore=\"test-logstore-1\" access-key-id=\"<yourAccessKeyId>\" access-key-secret=\"<yourAccessKeySecret>\"] %msg%\n") 
     *.* action(type="omfwd" protocol="tcp" target="cn-hangzhou.log.aliyuncs.com" port="10009" template="LogServiceFormat")

3. Restart the rsyslog utility.
   Run the sudo service rsyslog restart, sudo /etc/init.d/syslog-ng restart, or systemctl restart rsyslog command to restart the rsyslog utility.
4. Run the logger command to generate test logs.
   For example, run the logger hello world! command to generate logs.

Example 2: Use the syslog-ng utility to upload syslog logs to Log Service

1. Open the syslog-ng configuration file.
   The default path of the syslog-ng configuration file is /etc/syslog-ng/syslog-ng.conf.
2. Configure the following fields based on your syslog-ng version, and add the configurations to the end of your syslog-ng configuration file:

   ### Syslog-ng Logging Config for LogService ### 
   template LogServiceFormat { 
       template("<${PRI}>1 ${ISODATE} ${HOST:--} ${PROGRAM:--} ${PID:--} ${MSGID:--} [logservice project=\"test-project-1\" logstore=\"test-logstore-1\" access-key-id=\"<yourAccessKeyId>\" access-key-secret=\"<yourAccessKeySecret>\"] $MSG\n"); template_escape(no); 
   }; 
   destination d_logservice{ 
        tcp("cn-hangzhou.log.aliyuncs.com" port(10009) 
        tls(peer-verify(required-untrusted)) 
        template(LogServiceFormat)); 
   }; 
   log { 
        source(s_src); # default use s_src 
        destination(d_logservice); 
   }; 
   ### END Syslog-ng Logging Config for LogService ###

3. Restarts the rsyslog utility.
   Run the sudo /etc/init.d/syslog-ng restart, sudo service syslog-ng restart, or sudo systemctl restart syslog-ng command to restart the rsyslog utility.
4. Run the logger command to generate test logs.
   For example, run the logger hello world! command to generate logs.

Sample log entries

FAQ

• How can I send syslog logs to Log Service?

  You can run an Ncat command to simulate log uploads to Log Service. This way, you can check whether the network connection is normal and whether the AccessKey pair is authorized to send syslog logs. If Ncat is not installed on your server, you can run the sudo yum install nmap-ncat command to install Ncat.

  Note

  • The timestamp of syslog logs is in UTC+0, for example, 2019-03-28T03:00:15.003Z. In UTC+8, the timestamp is 2019-03-28T11:00:15.003.
  • Ncat commands cannot determine whether network connections are interrupted. After you run an Ncat command, you must enter the information to be sent and press the Enter key within 30 seconds.
  For example, you can run the following command to send a syslog log entry to Log Service. The project in Log Service is named test-project-1 and the Logstore is named test-logstore-1. The project resides in the China (Hangzhou) region. A RAM user that has the write permission is used to send the syslog log entry. The AccessKey ID of the RAM user is <yourAccessKeyId>, and the AccessKey secret is <yourAccessKeySecret>.

  [root@iZbp145dd9fccuidd7g**** ~]# ncat --ssl cn-hangzhou.log.aliyuncs.com 10009 
  <34>1 2019-03-28T03:00:15.003Z mymachine.example.com su - ID47 [logservice project="test-project-1" logstore="test-logstore-1" access-key-id="<yourAccessKeyId>" access-key-secret="<yourAccessKeySecret>"] this is a test message

  After you send the syslog log entry, you can preview the log entry in the Log Service console. For more information, see Preview logs.
• What can I do if a manual log upload fails?

  Troubleshoot the failure based on the error message. For more information, see Diagnose collection errors.

• How do I view rsyslog error logs?
  You can run the vim command to view the logs. By default, rsyslog error logs are stored in the /var/log/message directory.

  • Error log entry 1

    dlopen: /usr/lib64/rsyslog/lmnsd_gtls.so: cannot open shared object file: No such file or directory

    This error log entry is returned because the rsyslog-gnutls module is not installed. You can run the sudo apt-get install rsyslog-gnutls or sudo yum install rsyslog-gnutls command to install the module. Then restart the rsyslog utility.

  • Error log entry 2

    unexpected GnuTLS error -53 - this could be caused by a broken connection. GnuTLS reports:Error in the push function

    This error log entry is returned because the TCP connection is ended due to a long period of inactivity. You can ignore this error because rsyslog will re-establish the connection.

• How do I view rsyslog-ng error logs?

  You can run the systemctl status syslog-ng.service or journalctl-xe command to view the error logs. By default, rsyslog-ng error logs are stored in journal logs.

  If the following error log entry is returned, check whether the configuration file format is valid or whether configuration conflicts exist. For example, you cannot configure multiple internal() sources.

  Job for syslog-ng.service failed because the control process exited with error code. See "systemctl status syslog-ng.service" and "journalctl -xe" for details