,

Fixed issues

• Fixed an issue where container file descriptors were not released when a reader was restored from a checkpoint.

• Fixed an issue where feedback caused timeout events and log truncation.

Optimizations

• Enhanced the alerts for processor timeouts and zero-width regular expressions.

Fixed issues

• Fixed an issue where the rotator reader did not close files when a container stopped, which caused file descriptor leaks.

New features

• Added support for the hardened ECS metadata access mode.

Fixed issues

• Fixed a timeout issue that the Docker inspect command may cause.

• Fixed an issue where the rotator reader did not close files when a container stopped.

Fixed issues

• Fixed a crash in Golang caused by an incorrect null check on an interface.

• Fixed a crash that JsonLogFileReader may encounter due to a buffer overflow during a read operation.

• Fixed meaningless null characters that occurred when files were read.

Fixed issues

• Fixed a failure where Logtail collection could not automatically resume after the disk recovered from a full-disk state.

New features

• Added support for the CRI V1 interface for compatibility with containerd 2.0.

Fixed issues

• Updated the runc library to fix a file descriptor leak vulnerability in runc.

Optimizations

• Optimized the process of writing metric data to MetricStore.

Fixed issues

• Fixed an issue where the legacy standard output collection plug-in did not completely collect the last part of the logs when a container stopped.

Fixed issues

• Fixed an issue where directories were incorrectly concatenated when the path in a Linux collection configuration was the root directory.

• Fixed an issue where concurrent writes in container scenarios caused incorrect mount information in the container information.

• Fixed an issue where metric collection data was in an incorrect format.

• Fixed an issue where the metadata of some containers was missing during container collection.

Optimizations

• Added the FORCE_RELEASE_STOP_CONTAINER_FILE environment variable to forcibly release file handles when an application container stops.

Fixed issues

• Fixed an issue where the container status was incorrectly identified due to persistent volume mounts when container files were collected.

• Fixed an issue where the collection path was incorrectly identified due to multiple slashes in the path of a container collection configuration.

• Fixed an issue with how the C++ standard output log parsing module parsed the last line of a log.

• Fixed an issue where the enable_containerd_upper_dir_detect environment variable did not take effect in Containerd runtime scenarios.

New features

• The C++ standard output plug-in can now be configured by using environment variables.

• Added support for sending OTLP trace data to Simple Log Service.

Fixed issues

• Fixed an issue where container information was missing because the up command was run repeatedly in Docker Compose scenarios.

• Fixed an issue where self-monitoring CPU metrics were missing.

• Fixed a crash caused by the failure of the C++ standard output plug-in to parse the last line of a log.

New features

• Added container standard output collection implemented in C++.

• Upgraded self-monitoring. The C++ module and the Golang plug-in module now report self-monitoring data in a unified manner.

Fixed issues

• Fixed an issue where the binary log plug-in could not obtain the GTID when the GTID exceeded 1,024 bytes.

• Fixed a parsing failure in the native separator plug-in when a string contained double quotation marks.

Fixed issues

• Fixed the token expiration issue in the ENV-based management method that was introduced in version 2.0.7.

Optimizations

• Logtail now exits immediately when its resource usage exceeds the hard limit.

Fixed issues

• Fixed JSON truncation caused by escaped null bytes.

• Fixed a coredump caused by the use of the non-thread-safe gethostbyname method.

• Fixed an issue where tags were missing when OpenTelemetry parsed gauge metric data.

• Fixed an issue where readers that exceeded the reader queue length failed to be restored when too many files were rotated during checkpoint recovery. This further caused a new reader to read the checkpoint of an incorrect old reader during inode reuse, which led to truncation and duplicate collection.

• Fixed inaccurate GTIDs in the input_canal plug-in.

New features

• Added support for the rund architecture of Container Compute Service (ACS).

New features

• Added support for removing nanoseconds from the system time.

Optimizations

• Optimized the retry and fault tolerance mechanism for the containerd socket.

Fixed issues

• Fixed an issue introduced in version 2.0.4 where environment variable configurations did not take effect in Kubernetes scenarios.

• Fixed a panic in the plug-in module caused by the GetRealPath function in standard output scenarios.

• Fixed an issue where Shennong self-monitoring logs were missing when extension plug-ins were included.

• Fixed log truncation caused by read timeouts.

• Fixed missing metrics for the log splitting plug-in.

Optimizations

• Optimized the retry mechanism for container runtime detection and fixed inaccurate runtime identification.

• Added support for DNS cache when configurations are pulled from the config server.

• Changed the default timeout for multi-line splitting to 60 seconds.

• processor_gotime now supports nanosecond configuration.

Fixed issues

• Fixed an issue where the binary log plug-in could not obtain the GTID when the GTID exceeded 1,024 bytes.

• Fixed a parsing failure in the native separator plug-in when a string contained double quotation marks.

• Fixed a thread safety issue with the internal gethostbyname function.

Optimizations

• Added support for using the ALICLOUD_SLS_CLIENT_AUTH_VERSION environment variable to configure V4 authentication for the SDK that Logtail uses.

Optimizations

• Optimized the retry mechanism for container runtime detection.

Fixed issues

• Fixed an issue where the containerd runtime was inaccurately identified when the dockershim file existed.

• Fixed a coredump caused by *// in a file path.

• Fixed an issue where file paths were abnormally obtained when container standard output was collected.

• Fixed an issue where data was incompletely flushed when files were rotated.

Optimizations

• Optimized log printing for the sender module.

Fixed issues

• Ensured compatibility with duplicate keys in log content and preserved the correct key order.

• Fixed an issue where logs were still collected after the aliuid user ID file was deleted.

New features

• The Logtail-ds component now supports resource group configuration in ACK scenarios.

• Added the processor_rate_limit plug-in to the extension plug-ins.

Fixed issues

• Fixed an issue where the Logtail process could not exit gracefully after it collected historical data.

Optimizations

• Optimized the metric data and reporting logic of Golang plug-ins, which now report all metrics through the C++ metric module.

Fixed issues

• Fixed an issue where the Golang module was incorrectly loaded after the high-precision timestamp switch was turned on.

Fixed issues

• Fixed an exception in the Golang plug-in module caused by multithreading of tag data.

Optimizations

• The goprofile plug-in now uses the IP address of the machine in reported data.

• When native parsing fails and the raw log is retained, the raw log is retained only in the __raw_log__ field and no longer in the content field. This avoids data duplication.

Fixed issues

• Fixed a compatibility issue where time_key was not set to time by default in regex configurations.

• Fixed occasional crashes caused by using libcurl without setting CURLOPT_NOSIGNAL.

• Fixed an issue where fields were misaligned when the native separator parsing plug-in parsed logs that started with a space.

• Fixed an incorrect time zone handling issue when native plug-ins discarded timeout logs.

• Fixed an issue where the native JSON plug-in always incorrectly retained the original content field after it parsed any JSON that contained a content key.

• Fixed a memory leak in the native separator plug-in.

• Fixed duplicate logs caused by checkpoint dumping that occurred before directory registration.

• Fixed a compatibility issue where Apsara logs with comma-separated time formats could not be parsed.

• Fixed an issue where the container IP address obtained was sometimes empty when the pod network of a Kubernetes cluster was HostNetwork.

Fixed issues

• Resolved duplicate log collection introduced by the container memory fix.

• Fixed a plug-in crash caused by nil fields in the container information.

• Fixed an issue where ProcessorParseDelimiterNative parsed data that carried the next line.

• Fixed an issue where files could not be completely read under backpressure.

• Fixed a plug-in crash caused by a plugin_export panic.

• Fixed a crash caused by parsing Apsara-format logs.

• Fixed an issue where parsed data was concatenated when Apsara-format logs were parsed.

New features

• Added support for splitting lines based on a timeout.

• The Simple Log Service Flusher now supports nanosecond-level log time.

• Added a global host path blacklist.

• Added the processor_otel_metric trace parsing plug-in.

• Added support for adding resource tags when you create Logtail configurations by using environment variables.

Optimizations

• Cached logs that do not form a complete line to reduce file system read calls.

• Added support for using environment variables to control the log printing level.

• The skywalking plug-in now supports capturing the db.connection_string tag.

• Validated network interface controller (NIC) IP addresses to obtain a more accurate host IP address.

Fixed issues

• Resolved duplicate data collection that occurred when a statefulset with mounted volumes drifted to a different node.

• Fixed an issue where the name of the collected log file was incorrect after the inode of an empty file was reused.

• Fixed an issue where a container could not exit because a file was reopened based on a checkpoint.

• Fixed an issue where the time zone of Apsara log time could not be adjusted.

• Fixed an issue where the last line of a log may be incorrectly parsed in JSON mode if the line did not end with a line break.

• Fixed a JSON parsing exception that occurred when the beginning of the read data contained invalid JSON.

New features

• Added the input_command collection plug-in to collect the results of command execution.

• Added the processor_log_to_sls_metric processing plug-in to parse logs into metrics.

• Added a timeout configuration for the collection of disk-related metrics to avoid collection exceptions.

Optimizations

• The processor_json plug-in now supports parsing data in JSON array format.

• containerd container collection now supports custom rootfs paths and automatic path discovery.

• Optimized the reporting logic of the container metadata preview feature.

Fixed issues

• Fixed an issue where container standard output (stdout) could not be collected when the standard output file path was a symbolic link.

• Fixed an issue where a container could not exit because its FD was locked when only a container deletion event occurred.

• Fixed a nil panic in the service_go_profile plug-in.

• Fixed an issue where profile data metadata was inaccurate in container scenarios.

New features

• The Full-stack Monitoring application now supports collecting Kubernetes openkruise metric data.

Optimizations

• Added container metadata to Logtail profile data.

Fixed issues

• Fixed an issue where FDs were not released when a container exited.

• Fixed an Env configuration cache issue in Kubernetes scenarios.

New features

• Added the data_endpoint_policy startup parameter, which allows you to configure the policy that Logtail uses to switch the endpoint for accessing Simple Log Service.

• The Profiling feature now supports the goprofile pull mode.

• Added the processor_string_replace string replacement plug-in.

• Logtail now supports configuring a network proxy by using HTTP_PROXY.

Optimizations

• Optimized the performance of the processor_split_key_value plug-in and added multi-character quotes.

Fixed issues

• Fixed a disk usage metric statistics issue in the metric_system_v2 plug-in.

New features

• The HTTP input service now supports the Pyroscope protocol.

• Added support for reporting container information from the collection endpoint to improve the observability of container collection configurations.

Optimizations

• The ENV-based automatic collection method now supports more LogStore-related configuration parameters, such as the cold storage configuration.

Optimizations

• The prometheus scraping plug-in now supports Staleness data.

Fixed issues

• Fixed time zone issues, including an issue where the time zone adjustment option was ignored when the system time was used or when log time parsing failed.

• Fixed duplicate log collection caused by inode reuse.

• Fixed an issue where the Grok plug-in may not work when it parsed Chinese characters.

• Fixed the excessive memory usage of container discovery that was introduced in version 1.2.1.

New features

• Added support for ingesting OpenTelemetry protocol logs over HTTP.

• Added the processor_desensitize data masking plug-in.

Optimizations

• In container environments, the HTTPS protocol is now supported when you create Simple Log Service resources by using environment variables.

• In container environments, you can now select the LogStore specification when you create a LogStore by using a Custom Resource Definition (CRD).

• The offset of the content in the file can now be output when plug-ins are used to process data.

• Continuous context is now maintained by default when container standard output is collected.

• Optimized the memory usage of Prometheus data ingestion.

Fixed issues

• Fixed potential FD leaks and missed events in Docker environments.

• Fixed a file handle leak that occurred when Logtail collection configurations were updated.

• Fixed an issue where IP addresses were incorrectly resolved for special hostnames.

• Fixed duplicate file collection that occurred when multiple configuration paths had a parent-child directory relationship.

New features

• Added support for collecting SQL Server and PostgreSQL query results. For more information, see Collect SQL Server query results and Collect PostgreSQL query results.

• Added support for collecting JMX performance metrics.

• Added the log context aggregation plug-in (aggregator_context). For more information, see aggregators configuration.

  This makes features such as contextual query, __topic__ field extraction, and LiveTail available for logs that are processed by Logtail plug-ins.

• Added the Grok plug-in to extract log fields.

  You can set multiple Grok expressions to match logs in different formats.

Optimizations

• Added support for collecting the standard output of containers that exit within seconds.

Fixed issues

• Fixed the parsing of microsecond timestamps in Apsara-format logs.

New features

• Added the Logtail CSV processing plug-in.

• Added support for Layer 4 and Layer 7 network traffic analysis by using eBPF. The HTTP, MySQL, PgSQL, Redis, and DNS protocols are supported.

New features

• The netping plug-in now supports httping and DNS resolution latency.

New features

• Added the Skywalking Logging API.

Optimizations

• Added support for quickly releasing the file handles of stopped containerd containers.

Fixed issues

• Fixed an issue where the Kubernetes labels of containerd containers could not be matched.

New features

• When you collect text logs, you can parse high-precision time by using the extended configuration ("enable_precise_timestamp": true) or the processor_strptime plug-in.

Optimizations

• Optimized the rootfs detection mechanism in Kubernetes scenarios.

• Optimized the container runtime identification mechanism in Kubernetes scenarios.

Fixed issues

• Fixed an exception in the netping plug-in on Windows.

New features

• Logtail collection configurations now support environment variable substitution.

• Added the netping plug-in, which collects network ping data between a specified IP address and a destination IP address.

• The gotime plug-in can now convert the extracted log time to the timestamp format.

• When syslog logs are collected, the _client_ip_ field is added to indicate the IP address of the client that transmits the logs.

Optimizations

• Optimized the memory usage of container standard output stream collection.

New features

• The Prometheus plug-in now supports collecting Prometheus data from the same machine by using multiple Logtail collection configurations.

• Windows nodes in Container Service for Kubernetes now support add-on token authentication.

Fixed issues

• Fixed incorrect threadNum and fdNum metrics in the process collection plug-in on Linux.

• Fixed the ConfigurationDiscoveryService not implement error in the SkyWalking plug-in.

Fixed issues

• Fixed an issue where label matching based on regular expressions became invalid when container standard output was collected. This issue occurred in Logtail 1.0.27 and Logtail 1.0.28.

New features

• Added support for collecting SNMP protocol data.

• The SkyWalking V3 plug-in now supports filtering instance attributes.

• Added support for configuring the Accept Partial Fields parameter in separator mode.

Fixed issues

• Fixed incorrect Span IDs in the SkyWalking V2 plug-in.

New features

• Added the processor_regex plug-in.

• Added support for multi-region configuration management.

Optimizations

• Optimized host metric collection to support IO Counter metrics.

Fixed issues

• Fixed an issue where the service_http_server plug-in did not release UNIX connections.

• Fixed a conflict that occurred when Logtail ran multiple collection configurations of the metric_meta_kubernetes plug-in at the same time.

New features

• Added support for collecting process metrics.

• Added the collection of file handles and the TCP protocol when host metrics are collected.

• Added support for collecting the metadata of Kubernetes clusters.

• Added support for collecting the metadata of hosts.

• Added a gRPC output plug-in.

• Added support for Kubernetes cluster semantic recognition when container logs are collected.

• Added support for collecting trace data from SkyWalking V2.

• Added support for running the input_canal plug-in on the Windows i386 platform.

Optimizations

• Improved the accuracy of host metric collection in container environments.

Optimizations

• Strengthened the handling logic for cases where the readdir API of the file system returns inaccurate metadata.

Fixed issues

• Fixed a potential crash that occurred when historical data was imported.

Fixed issues

• Fixed an issue where the data that Logtail sent immediately after startup did not carry the custom ID.

Fixed issues

• Fixed an issue where Logtail may stop reporting status data, which is not user data, to Simple Log Service when the network was interrupted in Global Accelerator mode.

New features

• Added the exactly_once_concurrency configuration item, which allows Logtail to record fine-grained checkpoint information at the file level on the local disk. For more information, see Logtail configurations (Legacy).

• Added the enable_log_time_auto_adjust configuration item, which allows the log time to adapt to the local time of the server. For more information, see Set Logtail startup parameters.

• Added the enable_log_position_meta configuration item, which adds the metadata of the original file of a log to the log. For more information, see Logtail configurations (Legacy).

• Added the specified_year configuration item, which uses the year in the current time or a specified year to complete the log time. For more information, see Logtail configurations (Legacy).

Fixed issues

• Fixed a parsing failure caused by incorrect handling of the P (partial) tag when container standard output was collected.

• Fixed an issue where the SpanID and ParentSpanID in Links were incorrect in cross-application scenarios of the service_skywalking_agent_v3 plug-in.

Optimizations

• Increased the timeout for requests to the container engine from 3 seconds to 30 seconds. Added the DOCKER_CLIENT_REQUEST_TIMEOUT environment variable, which sets the timeout for requests to the container engine.

Fixed issues

• Fixed a defect where the parent Span ID of the service_skywalking plug-in was incorrect.

• Fixed a defect where the logic that creates collection configurations based on environment variables may exit when the container engine was abnormal.

Fixed issues

• Fixed a data sending failure that occurred with a low probability when data was out of order.

New features

• Added support for collecting container data in containerd environments.

Optimizations

• Adjusted the scope of the net_err_stat metric in service logs so that it covers only sending errors caused by the network.

New features

• Added the net_err_stat metric to service logs, which records the number of sending errors that occurred in the last 1 minute, 5 minutes, and 15 minutes.

Optimizations

• Optimized the CPU overhead of container data collection scenarios.

New features

• Added support for installing the service_telegraf plug-in on demand at runtime. This feature is available only to ECS users.

Optimizations

• Optimized the service_telegraf plug-in to support multiple configurations on a single machine.

Optimizations

• Strictly limited the network types that Logtail can use.

New features

Added the service_telegraf plug-in, which collects metric data.

New features

• Blacklist filtering now supports multi-level matching, such as /path/**/log.

Optimizations

• Optimized the policy for obtaining the local IP address. If the original policy fails, the first IP address in the list is obtained.

New features

• Added the metric_system_v2 host status data plug-in.

• Added the ALIYUN_LOGTAIL_MAX_DOCKER_CONFIG_UPDATE_TIMES environment variable and its corresponding max_docker_config_update_times parameter, which apply to scenarios where short-lived Jobs are frequently created in Kubernetes environments.

Optimizations

• Optimized the performance, in terms of CPU overhead, of container collection scenarios that have many collection configurations.

Fixed issues

• Fixed an issue where the processor_split_log_string plug-in occasionally generated empty lines.

New features

• Added support for custom time field names in full regex mode.

• Added the KeepSourceIfParseError parameter to the processor_json, processor_regex, and processor_split_char plug-ins to retain raw data when parsing fails. For more information, see Use Logtail plug-ins to process data.

New features

• Added the processor_encrypt encryption plug-in.

New features

• Added an HTTP Probe to support Kubernetes health checks.

Fixed issues

• Fixed a core dump caused by libcurl in some environments.

• Fixed an issue where the libidn library was missing when Logtail was installed on CentOS 8.

New features

• Added the IgnoreFirstConnector parameter to the processor_json plug-in.

• Added support for millisecond and microsecond timestamps (%s).

• Added support for loading multiple local Logtail configurations from /etc/ilogtail/config.d/.

• Added support for loading multiple local user configurations from /etc/ilogtail/user_config.d/.

• Added the processor_split_key_value and processor_strptime processing plug-ins.

• Added the oas_connect_timeout and oas_request_timeout parameters to support scenarios with slow networks.

Optimizations

• Removed the restriction on inputs in hybrid configurations (file+plugin).

New features

• Added Kubernetes-level filtering for the collection of Docker standard output and files.

Optimizations

• Optimized concurrent contention between LogStores in the same region when network conditions are poor.

Fixed issues

• Fixed a checkpoint loss issue that occurred with a low probability due to an error in the file opening logic.

New features

• Added a parameter that configures the tail size of the first collection.

Optimizations

• Optimized the logic for obtaining container metadata to reduce the overall impact of abnormal containers.

Fixed issues

• Fixed a memory leak of docker_stdout in complex environments.

• Fixed incomplete support for millisecond timestamps in JSON mode.

New features

• Added support for collecting containerd logs.

Fixed issues

• Fixed a checkpoint loss issue for rotated files that occurred with an extremely low probability.

• Fixed an issue where the local collection configuration file /etc/ilogtail/user_config.d was not loaded when the /usr/local/ilogtail/user_log_config.json file did not exist.

New features

• Added support for configuring working_ip and working_hostname by using environment variables.

• Added the force_quit_read_timeout parameter, which sets the timeout for a forced exit when events cannot be read due to a persistent block.

• Added support for passing tags such as path and topic to plug-ins.

• Added the aggregator_shardhash plug-in, which allows you to set shardhash within the plug-in.

• Added the processor_gotime, processor_rename, processor_add_fields, processor_json, and processor_packjson processing plug-ins. For more information, see Use Logtail plug-ins to process data.

• Updated LogtailInsight and added the progress viewing feature. You must set mark_offset_global_flag or customized_fields.mark_offset.

Optimizations

• Optimized the high memory usage of Journal when it runs for a long time by releasing memory as early as possible.

• Optimized the interval at which configurations are obtained for the first time when no local configuration exists.

Fixed issues

• Fixed potential duplicate collection when multiple Logtail configurations were used.

• Fixed an issue where millisecond and microsecond timestamps did not support JSON int64.

New features

• Added support for custom static topic settings.

• Added support for blacklist filtering.

• Added the EnableEventMeta parameter to the service_canal plug-in to collect the header information of MySQL binary logs.

Optimizations

• Optimized the stop mechanism of the plug-in system.

Fixed issues

• Fixed a potential memory leak with GBK logs.

New features

• Added support for collecting Docker events. For more information, see Collect Docker events.

• Added support for collecting Systemd Journal logs. For more information, see Collect Systemd Journal logs.

• Added the processor_pick_key and processor_drop_last_key processing plug-ins.

Optimizations

• Optimized the memory usage of container log collection and plug-in collection.

• Optimized the performance of collecting multi-line logs from container standard output.

New features

• Added support for automatically creating resources related to Kubernetes audit logs.

• Added support for configuring startup parameters, such as CPU, memory, and sending concurrency, by using environment variables.

• Added support for configuring custom tag uploads by using environment variables.

• Sidecar mode now supports automatic configuration creation. For more information, see Collect container text logs using a sidecar CRD.

Optimizations

• The aliuid file is now saved automatically to a local file.

Fixed issues

• Fixed a crash that occurred with an extremely low probability when container files were collected.

• Fixed an issue where IncludeLabel did not take effect in Kubernetes for configurations that were created by using environment variables.

New features

• Added support for GTID mode when MySQL binary logs are collected. This mode is automatically enabled when MySQL binary logs are collected.

• Added support for specifying wildcards in the file names for historical data import.

• Kubernetes now supports the automatic creation of index configurations.

Optimizations

• When line splitting fails, discardUnMatch is checked and the logs that failed to be split are reported.

• Sending is now retried automatically when an unknown send error occurs, which prevents data loss in extremely rare cases, such as when a sent data packet is tampered with in transit.

New features

• Added support for wildcard mode when historical data is imported.

• Added the default_tail_limit_kb startup configuration item, which configures the skip size for the first collection of a file. The default value is 1,024 KB.

• Added the batch_send_seconds collection configuration item, which configures the interval at which data packets are sent.

• Added the batch_send_bytes collection configuration item, which configures the size of data packets.

Optimizations

• When container standard output is collected, logs that are split by the Docker Engine are automatically merged.

New features

• Added support for configuring log collection by using environment variables.

• Added support for collecting the metadata in MySQL binary logs. The _filename_ and _offset_ log fields are added.

• The installation script now supports automatic parameter selection in a VPC.

• Added support for the Global Accelerator installation mode. For more information, see Step 2: Configure Logtail collection acceleration.

Optimizations

• When MySQL binary logs are collected, filename and offset information can now be collected.

Fixed issues

• Fixed an exception that occurred with a certain probability when multi-line mode was used to collect container standard output.

Optimizations

• Upgraded the collection method for container standard output.

New features

• Added a frequency limit for updating container file collection configurations.

Fixed issues

• Fixed a socket fd leak that occurred with an extremely low probability.

New features

• Added the Logtail Lumberjack plug-in, which collects data from Logstash and Beats data sources. For more information, see Collect data from Beats and Logstash data sources.

• Added the inotify blacklist feature.

Fixed issues

• Fixed inconsistent parameters in earlier installation packages.

• Fixed an issue where the OS version could not be correctly obtained when Logtail was installed on some systems.

New features

• Added support for collecting host monitoring data.

• Added support for collecting Redis monitoring data.

• Added support for collecting DDL (data definition language) statements in MySQL binary logs.

• Added support for filtering by docker ENV (environment) when container standard output and container files are collected.

Fixed issues

• Ensured compatibility with data collection from MySQL tables that have no primary key.

• Addressed event loss caused by an unstable container engine subscription channel in container collection mode.

New features

• Added the multi-line collection mode for container standard output. For more information, see Examples of Logtail configurations for multi-line logs.

New features

• Added support for Docker and Kubernetes deployment solutions.

• Added support for collecting container standard output and container files. For more information, see Collect container standard output using a DaemonSet in the console and Collect container text logs using a DaemonSet in the console.

New features

• Added the processor_geoip plug-in.

New features

• Added support for collecting MySQL binary logs, MySQL query results, and HTTP data. For more information, see Use Logtail plug-ins to collect data.

• Added support for combined parsing configurations: regex mode, anchor mode, separator mode, and filters.