If you use consumer groups to consume log data, you do not need to consider factors such as Log Service implementation, load balancing among consumers, or failovers that may occur. You can focus on business logic during log data consumption.

Prerequisites

An SDK development environment is set up. For more information, see Overview of Log Service SDKs.

Terms

Term Description
consumer group You can use consumer groups to consume data in Log Service. A consumer group consists of multiple consumers. All consumers in a consumer group consume data in a Logstore and do not repeatedly consume data. You can create up to 30 consumer groups for a Logstore.
consumer The consumers in a consumer group consume data. The names of the consumers in a consumer group must be unique.
Logstore A Logstore in Log Service is a unit that is used to collect, store, and query log data. For more information, see Logstore.
shard A shard is used to control the read and write capacity of a Logstore. In Log Service, data is stored in shards. Each shard has an MD5 hash range, and each range is a left-closed, right-open interval. Each range does not overlap with the ranges of other shards. Each range must be within the entire MD5 hash range [00000000000000000000000000000000,ffffffffffffffffffffffffffffffff). For more information, see Shard.
checkpoint A checkpoint is the position at which a program stops consuming data. If the program is restarted, the program consumes data from the last checkpoint.
A Logstore has multiple shards. Log Service allocates shards to the consumers in a consumer group based on the following rules:
  • Each shard can be allocated to only one consumer.
  • A consumer can consume data from multiple shards.

After a new consumer is added to a consumer group, shards that are allocated to the consumer group are reallocated to each consumer for load balancing. The shards are reallocated based on the preceding rules.

If you use consumer groups to consume data, Log Service stores checkpoints. Consumers can resume data consumption from a checkpoint without repeatedly consuming data after a program is recovered from a fault.

Procedure

You can use Log Service SDK for Java, C++, Python, or Go to create consumer groups and consume data. In the following example, Log Service SDK for Java is used.

  1. Add Maven dependencies.
    Open the pom.xml file in the root directory of the Java project and add the following code:
    <dependency>
  <groupId>com.google.protobuf</groupId>
  <artifactId>protobuf-java</artifactId>
  <version>2.5.0</version>
</dependency>
<dependency>
  <groupId>com.aliyun.openservices</groupId>
  <artifactId>loghub-client-lib</artifactId>
  <version>0.6.33</version>
</dependency>
  2. Create a Main.java file. 
    import com.aliyun.openservices.loghub.client.ClientWorker;
import com.aliyun.openservices.loghub.client.config.LogHubConfig;
import com.aliyun.openservices.loghub.client.exceptions.LogHubClientWorkerException;

public class Main {
    // The Log Service endpoint. Enter the endpoint based on your business requirements. For more information, see Endpoints. 
    private static String sEndpoint = "cn-hangzhou.log.aliyuncs.com";
    // The name of the Log Service project. Enter the name based on your business requirements. 
    private static String sProject = "ali-cn-hangzhou-sls-admin";
    // The name of the Logstore. Enter the name based on your business requirements. 
    private static String sLogstore = "sls_operation_log";
    // The name of the consumer group. Enter the name based on your business requirements. 
    private static String sConsumerGroup = "consumerGroupX";
    // The AccessKey ID and AccessKey secret of the user that consumes data. Enter the AccessKey ID and AccessKey secret based on your business requirements. For more information, see AccessKey pair. 
    private static String sAccessKeyId = "";
    private static String sAccessKey = "";

    public static void main(String[] args) throws LogHubClientWorkerException, InterruptedException {
        // consumer_1 specifies the name of a consumer. The name of each consumer in a consumer group must be unique. If different consumers start processes on different machines to consume data in a Logstore, you can use the machine IP addresses to identify each consumer. 
        // maxFetchLogGroupSize specifies the maximum number of log groups that you can obtain from Log Service at a time. Retain the default value. You can use config.setMaxFetchLogGroupSize(100); to change the maximum number. The maximum number must be within the range (0,1000]. 
        LogHubConfig config = new LogHubConfig(sConsumerGroup, "consumer_1", sEndpoint, sProject, sLogstore, sAccessKeyId, sAccessKey, LogHubConfig.ConsumePosition.BEGIN_CURSOR,1000);
        ClientWorker worker = new ClientWorker(new SampleLogHubProcessorFactory(), config);
        Thread thread = new Thread(worker);
        // After the Thread class runs, the ClientWorker class automatically runs and implements the Runnable interface. 
        thread.start();
        Thread.sleep(60 * 60 * 1000);
        // The shutdown function of the worker instance is called to stop the consumer. The associated thread is automatically stopped. 
        worker.shutdown();
        // Multiple asynchronous tasks are generated when the ClientWorker class is running. To ensure that all running tasks securely stop after shutdown, we recommend that you set Thread.sleep to 30 seconds. 
        Thread.sleep(30 * 1000);
    }
}
  3. Create a SampleLogHubProcessor.java file. 
    import com.aliyun.openservices.log.common.FastLog;
import com.aliyun.openservices.log.common.FastLogContent;
import com.aliyun.openservices.log.common.FastLogGroup;
import com.aliyun.openservices.log.common.FastLogTag;
import com.aliyun.openservices.log.common.LogGroupData;
import com.aliyun.openservices.loghub.client.ILogHubCheckPointTracker;
import com.aliyun.openservices.loghub.client.exceptions.LogHubCheckPointException;
import com.aliyun.openservices.loghub.client.interfaces.ILogHubProcessor;
import com.aliyun.openservices.loghub.client.interfaces.ILogHubProcessorFactory;

import java.util.List;

public class SampleLogHubProcessor implements ILogHubProcessor {
    private int shardId;
    // The time when the last checkpoint is saved. 
    private long mLastCheckTime = 0;

    public void initialize(int shardId) {
        this.shardId = shardId;
    }

    // The main logic of data consumption. You must include the code to handle all exceptions that may occur during data consumption. 
    public String process(List<LogGroupData> logGroups,
                          ILogHubCheckPointTracker checkPointTracker) {
        // The obtained data is displayed. 
        for (LogGroupData logGroup : logGroups) {
            FastLogGroup flg = logGroup.GetFastLogGroup();
            System.out.println(String.format("\tcategory\t:\t%s\n\tsource\t:\t%s\n\ttopic\t:\t%s\n\tmachineUUID\t:\t%s",
                    flg.getCategory(), flg.getSource(), flg.getTopic(), flg.getMachineUUID()));
            System.out.println("Tags");
            for (int tagIdx = 0; tagIdx < flg.getLogTagsCount(); ++tagIdx) {
                FastLogTag logtag = flg.getLogTags(tagIdx);
                System.out.println(String.format("\t%s\t:\t%s", logtag.getKey(), logtag.getValue()));
            }
            for (int lIdx = 0; lIdx < flg.getLogsCount(); ++lIdx) {
                FastLog log = flg.getLogs(lIdx);
                System.out.println("--------\nLog: " + lIdx + ", time: " + log.getTime() + ", GetContentCount: " + log.getContentsCount());
                for (int cIdx = 0; cIdx < log.getContentsCount(); ++cIdx) {
                    FastLogContent content = log.getContents(cIdx);
                    System.out.println(content.getKey() + "\t:\t" + content.getValue());
                }
            }
        }
        long curTime = System.currentTimeMillis();
        // A checkpoint is written to Log Service at an interval of 30 seconds. If the worker instance unexpectedly stops within 30 seconds, a newly started worker instance obtains consumption data from the last checkpoint. A small amount of duplicate data may be obtained. 
        if (curTime - mLastCheckTime > 30 * 1000) {
            try {
                // If true is passed to saveCheckPoint, checkpoints are immediately synchronized to Log Service. If false is passed to saveCheckPoint, checkpoints are cached on your computer. By default, checkpoints are synchronized to Log Service at an interval of 60 seconds. 
                checkPointTracker.saveCheckPoint(true);
            } catch (LogHubCheckPointException e) {
                e.printStackTrace();
            }
            mLastCheckTime = curTime;
        }
        return null;
    }

    // The shutdown function of the worker instance is called. You can manage the checkpoints. 
    public void shutdown(ILogHubCheckPointTracker checkPointTracker) {
        // The checkpoints are immediately saved to Log Service. 
        try {
            checkPointTracker.saveCheckPoint(true);
        } catch (LogHubCheckPointException e) {
            e.printStackTrace();
        }
    }
}

class SampleLogHubProcessorFactory implements ILogHubProcessorFactory {
    public ILogHubProcessor generatorProcessor() {
        // A consumer is generated. 
        return new SampleLogHubProcessor();
    }
}
    For more information, see Java, C++, Python, and Go.
  4. Run Main.java.
    In the following example, NGINX logs are consumed and the consumption result is displayed.
    :    GET
request_uri    :    /request/path-3/file-7
status    :    200
body_bytes_sent    :    3820
host    :    www.example.com
request_time    :    43
request_length    :    1987
http_user_agent    :    Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2228.0 Safari/537.36
http_referer    :    www.example.com
http_x_forwarded_for    :    192.168.10.196
upstream_response_time    :    0.02
--------
Log: 158, time: 1635629778, GetContentCount: 14
......
    category    :    null
    source    :    127.0.0.1
    topic    :    nginx_access_log
    machineUUID    :    null
Tags
    __receive_time__    :    1635629815
--------
Log: 0, time: 1635629788, GetContentCount: 14
......
    category    :    null
    source    :    127.0.0.1
    topic    :    nginx_access_log
    machineUUID    :    null
Tags
    __receive_time__    :    1635629877
--------
......

View the status of a consumer group in the Log Service console

  1. Log on to the Log Service console.
  2. In the Projects section, click the name of the project that you want to manage.
  3. Click the Show icon icon of the Logstore that you want to manage and click Data Consumption.
  4. Click the consumer group whose status you can view. Then, you can view the data consumption progress of each shard.

View the status of a consumer group by using code

You can view the data consumption progress by using an SDK. In the following example, Log Service SDK for Java is used, and ConsumerGroupTest.java is executed. 
import java.util.List;
import com.aliyun.openservices.log.Client;
import com.aliyun.openservices.log.common.Consts.CursorMode;
import com.aliyun.openservices.log.common.ConsumerGroup;
import com.aliyun.openservices.log.common.ConsumerGroupShardCheckPoint;
import com.aliyun.openservices.log.exception.LogException;
public class ConsumerGroupTest {
    static String endpoint = "";
    static String project = "";
    static String logstore = "";
    static String accesskeyId = "";
    static String accesskey = "";
    public static void main(String[] args) throws LogException {
        Client client = new Client(endpoint, accesskeyId, accesskey);
        // Obtain all consumer groups that are created for the Logstore. If no consumer groups exist, an empty string is returned. 
        List<ConsumerGroup> consumerGroups = client.ListConsumerGroup(project, logstore).GetConsumerGroups();
        for(ConsumerGroup c: consumerGroups){
            // Display the properties of each consumer group, including the name, heartbeat timeout period, and whether log data is consumed in order. 
            System.out.println("Name: " + c.getConsumerGroupName());
            System.out.println("Heartbeat timeout period: " + c.getTimeout());
            System.out.println("Ordered consumption: " + c.isInOrder());
            for(ConsumerGroupShardCheckPoint cp: client.GetCheckPoint(project, logstore, c.getConsumerGroupName()).GetCheckPoints()){
                System.out.println("shard: " + cp.getShard());
                // The consumption time. The value is a long integer with microsecond precision. 
                System.out.println("The time when data is last consumed: " + cp.getUpdateTime());
                System.out.println("Consumer name: " + cp.getConsumer());
                String consumerPrg = "";
                if(cp.getCheckPoint().isEmpty())
                    consumerPrg = "Consumption is not started";
                else{
                    // The UNIX timestamp. Unit: seconds. Format the output value of the timestamp. 
                    try{
                        int prg = client.GetPrevCursorTime(project, logstore, cp.getShard(), cp.getCheckPoint()).GetCursorTime();
                        consumerPrg = "" + prg;
                    }
                    catch(LogException e){
                        if(e.GetErrorCode() == "InvalidCursor")
                            consumerPrg = "Invalid. The time when data is last consumed is out of the retention period of the data.";
                        else{
                            //internal server error
                            throw e;
                        }
                    }
                }
                System.out.println("Consumption progress: " + consumerPrg);
                String endCursor = client.GetCursor(project, logstore, cp.getShard(), CursorMode.END).GetCursor();
                int endPrg = 0;
                try{
                    endPrg = client.GetPrevCursorTime(project, logstore, cp.getShard(), endCursor).GetCursorTime();
                }
                catch(LogException e){
                    //do nothing
                }
                // The UNIX timestamp. Unit: seconds. Format the output value of the timestamp. 
                System.out.println("The time when the last data record is received: " + endPrg);
            }
        }
    }
}
The following information is returned:
Name: etl-6cac01c571d5a4b933649c04a7ba215b
Heartbeat timeout period: 60
Ordered consumption: false
shard: 0
The time when data is last consumed: 1639555453575211
Consumer name: etl-356464787983a3d17086a9797e3d5f0e6959b066-256521
Consumption progress: 1639555453
The time when the last data record is received: 1639555453
shard: 1
The time when data is last consumed: 1639555392071328
Consumer name: etl-356464787983a3d17086a9797e3d5f0e6959b066-256521
Consumption progress: 1639555391
The time when the last data record is received: 1639555391
Name: etl-2bd3fdfdd63595d56b1ac24393bf5991
Heartbeat timeout period: 60
Ordered consumption: false
shard: 0
The time when data is last consumed: 1639555453256773
Consumer name: etl-532719dd2f198b3878ee3c6dfc80aeffb39ee48b-061390
Consumption progress: 1639555453
The time when the last data record is received: 1639555453
shard: 1
The time when data is last consumed: 1639555392066234
Consumer name: etl-532719dd2f198b3878ee3c6dfc80aeffb39ee48b-061390
Consumption progress: 1639555391
The time when the last data record is received: 1639555391
Name: consumerGroupX
Heartbeat timeout period: 60
Ordered consumption: false
shard: 0
The time when data is last consumed: 1639555434142879
Consumer name: consumer_1
Consumption progress: 1635615029
The time when the last data record is received: 1639555453
shard: 1
The time when data is last consumed: 1639555437976929
Consumer name: consumer_1
Consumption progress: 1635616802
The time when the last data record is received: 1639555391

What to do next

  • Configure Log4j for troubleshooting.
    We recommend that you configure Log4j for the consumer program to return error messages when consumer groups encounter exceptions. This helps you troubleshoot the errors. The following code shows a log4j.properties configuration file that is commonly used: 
    log4j.rootLogger = info,stdout
log4j.appender.stdout = org.apache.log4j.ConsoleAppender
log4j.appender.stdout.Target = System.out
log4j.appender.stdout.layout = org.apache.log4j.PatternLayout
log4j.appender.stdout.layout.ConversionPattern = [%-5p] %d{yyyy-MM-dd HH:mm:ss,SSS} method:%l%n%m%n
    After you configure Log4j, you can view information about exceptions when you run the consumer program. The following code shows information about an exception: 
    [WARN ] 2018-03-14 12:01:52,747 method:com.aliyun.openservices.loghub.client.LogHubConsumer.sampleLogError(LogHubConsumer.java:159)
com.aliyun.openservices.log.exception.LogException: Invalid loggroup count, (0,1000]
  • Use a consumer group to consume data that is generated from a certain point in time. 
    // consumerStartTimeInSeconds specifies a point in time. The data that is generated after the point in time is consumed. 
public LogHubConfig(String consumerGroupName, 
                      String consumerName, 
                      String loghubEndPoint,
                      String project, String logStore,
                      String accessId, String accessKey,
                      int consumerStartTimeInSeconds);

// position is an enumeration variable. LogHubConfig.ConsumePosition.BEGIN_CURSOR specifies that the consumption starts from the earliest data. LogHubConfig.ConsumePosition.END_CURSOR specifies that the consumption starts from the latest data. 
public LogHubConfig(String consumerGroupName, 
                      String consumerName, 
                      String loghubEndPoint,
                      String project, String logStore,
                      String accessId, String accessKey,
                      ConsumePosition position);
    Note
    • You can use different constructors based on your business requirements.
    • If a checkpoint is stored on Log Service, data consumption starts from the checkpoint.
    • When Log Service consumes data, a checkpoint is preferentially used to start data consumption. If you want to specify a point in time for Log Service to start data consumption, make sure that the value of consumerStartTimeInSeconds falls within the time-to-live (TTL) period. Otherwise, Log Service cannot consume data based your configurations.
  • Reset a checkpoint. 
    public static void updateCheckpoint() throws Exception {
        Client client = new Client(host, accessId, accessKey);
        long timestamp = Timestamp.valueOf("2017-11-15 00:00:00").getTime() / 1000;
        ListShardResponse response = client.ListShard(new ListShardRequest(project, logStore));
        for (Shard shard : response.GetShards()) {
            int shardId = shard.GetShardId();
            String cursor = client.GetCursor(project, logStore, shardId, timestamp).GetCursor();
            client.UpdateCheckPoint(project, logStore, consumerGroup, shardId, cursor);
        }
    }

Authorize a RAM user to access consumer groups

Before you can use a RAM user to access consumer groups, you must grant the required permissions to the RAM user. For more information, see Step 2: Grant permissions to the RAM user.

The following table describes the actions that the RAM user is authorized to perform.
Action Description Resource
log:GetCursorOrData (GetCursor and PullLogs) Queries cursors based on the time when logs are generated. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}
log:CreateConsumerGroup Creates a consumer group for a specified Logstore. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}/consumergroup/${consumerGroupName}
log:ListConsumerGroup Queries all consumer groups of a specified Logstore. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}/consumergroup/*
log:ConsumerGroupUpdateCheckPoint Updates the checkpoint for a shard that is allocated to a specified consumer group. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}/consumergroup/${consumerGroupName}
log:ConsumerGroupHeartBeat Sends a heartbeat message for a specified consumer to Log Service. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}/consumergroup/${consumerGroupName}
log:UpdateConsumerGroup Modifies the attributes of a specified consumer group. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}/consumergroup/${consumerGroupName}
log:GetConsumerGroupCheckPoint Queries the checkpoints for one or all shards that are allocated to a specified consumer group. acs:log:${regionName}:${projectOwnerAliUid}:project/${projectName}/logstore/${logstoreName}/consumergroup/${consumerGroupName}
For example, the project-test project resides in the China (Hangzhou) region. The ID of the Alibaba Cloud account to which the project belongs is 174649****602745. The name of the Logstore from which you want to consume log data is logstore-test, and the name of the consumer group that you want to use is consumergroup-test. To allow a RAM user to access the consumer group, you must grant the following permissions to the RAM user: 
{
  "Version": "1",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "log:GetCursorOrData"
      ],
      "Resource": "acs:log:cn-hangzhou:174649****602745:project/project-test/logstore/logstore-test"
    },
    {
      "Effect": "Allow",
      "Action": [
        "log:CreateConsumerGroup",
        "log:ListConsumerGroup"
      ],
      "Resource": "acs:log:cn-hangzhou:174649****602745:project/project-test/logstore/logstore-test/consumergroup/*"
    },
    {
      "Effect": "Allow",
      "Action": [
        "log:ConsumerGroupUpdateCheckPoint",
        "log:ConsumerGroupHeartBeat",
        "log:UpdateConsumerGroup",
        "log:GetConsumerGroupCheckPoint"
      ],
      "Resource": "acs:log:cn-hangzhou:174649****602745:project/project-test/logstore/logstore-test/consumergroup/consumergroup-test"
    }
  ]
}