Release notes

Agora Cloud Recording is an add-on service to record and save video calls, voice calls, and interactive streaming to your cloud storage.

Agora Cloud Recording is compatible with Video SDK for:

• Android, iOS, macOS, and Windows - v1.7.0 or later
• Web - v1.12 or later

This page provides the release notes for:

• Cloud Recording
• Notifications

This release adds the following improvement:

To improve the security of Cloud Recording, this release adds optional stsToken and stsExpiration fields to the storageConfig of the start and update requests. You need to ensure that the validity period of stsToken and stsExpiration is long enough, so that Cloud Recording can successfully upload files to a third-party cloud storage. For details, see and API reference.

This release adds the following improvements:

• Adds support for the following regions of Alibaba Cloud storage:

  • 23: CN_Nanjing
  • 24: CN_Fuzhou
  • 25: CN_Wulanchabu
  • 26: CN_Northeast_2
  • 27: CN_Southeast_7

• Updates the Agora transcoding script to reduce background noise.

This release adds the following improvements:

• Improves player compatibility for MPD files generated in individual recording standard mode, see Recording configuration.
• Optimizes the Agora Format Converter Script for individual recording, you can refer to Merge audio and video files to get the latest Agora Format Converter Script.

This release supports the postpone transcoding feature globally. You need to set the scene field to 2 (postpone transcoding) when calling the acquire. After setting this scene, the recording service transcodes the recorded file within 24 hours after recording to generate an MP4 file, and uploads the MP4 file to the third-party cloud storage you specify. Please note that the region of the third-party cloud storage must be consistent with the region set when calling acquire. For details, see Individual recording.

This release adds support for Amazon S3 in some regions. See Cloud storage configuration.

This release adds the following new feature and API changes:

Mobile web page mode for web page recording

This release adds mobile web page mode for web page recording. The API changes are as follows:

• Adds the mobile field in extensionServiceConfig. After setting mobile to true, the recording service uses the mobile web page mode to record the current web page.
• The value range of the videoWidth and videoHeight fields in extensionServiceConfig is changed to [240, 1920].

Transcoder script for high availability mechanism

In composite recording mode, Agora provides a transcoder script for merging the old M3U8 file with the new M3U8 file generated after the service enables the high availability mechanism into an MP4 file. For details, see Manage Recorded Files.

Individual Audio Non-transcoding Recording

In order to meet the diverse recording needs of users, this release adds the streamMode parameter in the start method to support Individual Audio Non-transcoding Recording.

Postpone Audio Mixing For voice chat scenarios, this release adds the appsCollection and transcodeOptions parameters in the start method to support Postpone Audio Mixing. If you enable this function, the recording service merges and transcodes the recorded files of all user IDs in the specified channel within 24 hours after the recording ends to generate an MP3/M4A/AAC file, and then uploads the recording file to your specified third-party cloud storage.

A Combined audio and video file after individual recording

When calling start, set the streamMode field of the recordingConfig parameter to standard, you can get a M3U8 audio index file, a M3U8 video index file and a combined M3U8 audio and video index file. If VP9 encoding is used on the web side, a combined MPD audio and video index file are generated. You can play the combined audio and video index file directly without using script. Agora recommends that you use this mode.

Support for the third-party cloud storage service encrypting and labeling recording files

This release adds the extensionParams field to support the third-party cloud storage service to encrypt and tag uploaded recording files. This feature is applicable to Amazon S3 and Alibaba Cloud OSS.

This release adds support for Alibaba Cloud (vendor is 2) in the following regions:

• 19：AP_Southeast_6
• 20：CN_Heyuan
• 21：CN_Guangzhou
• 22：CN_Chengdu

For details, see Alibaba Cloud Documentation.

This release fixed the following issue:

When a composite recording started, the user to be recorded did not send streams. When the user sent streams, the user could not display the customized background image or background color after they stopped sending streams or left the channel.

Support for AES-GCM2 encryption

For scenarios requiring high security, to guarantee the confidentiality, integrity, and authenticity of the data and to improve the computational efficiency of the data encryption, this release adds the following options in recordingConfig:

• AES_128_GCM2 and AES_256_GCM2 encryption modes in decryptionMode of recordingConfig. The new GCM encryption modes use a more secure KDF (Key Derivation Function) and support setting the key and salt.
• The salt parameter in recordingConfig to add the salt for the AES_128_GCM2 and AES_256_GCM2 encryption modes.

Push media stream to the CDN during a web page recording

Adds support for converting the content and audio of a web page into a media stream and pushing it to the CDN during the web page recording. See the start, update and query methods of the Agora Cloud Recording RESTful API for details.

Third-party cloud storage

Adds support for Google Cloud, Huawei Cloud and Baidu AI Cloud as third-party storage platforms.

Capture screenshots

This release increases the range of the captureInterval of snapshotConfig (in seconds) to [1,3600] from [5,3600] when using the Cloud Recording RESTful API to take screenshots of a video stream.

Output video resolution

This release increases the upper limit of the output video resolution generated by the web page recording from 1280 x 720 to 1920 x 1080. That is, the range of the videoWidth and videoHeight parameters of the web page recording in extensionServiceConfig is changed to [480, 1920].

The estimated maximum number of subscribed users

To support a larger estimated maximum number of subscribed users, this release adds the following enumeration values of subscribeUidGroup in recordingConfig:

• 4: 17 to 32 user IDs
• 5: 32 to 49 user IDs

The new server version requirements for transcoding recorded files are as follows:

• CentOS 7.0 and later x64
• Ubuntu 14.04 and later x64
• Debian 8.0 and later

Web page load timeout detection

In order to ensure the validity and completeness of page recording, this version supports page load timeout detection function. See Web Page Load Timeout Detection.

Pause the page recording

This release adds the onhold parameter to the start and update methods, which is used to pause the recording when the web page recording starts or is in progress.

Capture screenshots and recording in a recording process

This release adds support for recording and taking screenshots in a recording process. Only the recording fee is charged. See Capture Screenshots.

• The web_recorder_stopped callback adds error codes and error messages to indicate the reason for the abnormal stop of the web page recording. You can take corresponding measures based on the information.
• The update method adds the webRecordingConfig parameter, which is used to update related parameters during a web page recording.

This release includes the following new feature and API changes:

Screenshot event notification

This release adds the recorder_snapshot_file callback，which indicates that the screenshot is captured successfully, and returns the name of the screenshot file.

New regions for Amazon S3 cloud storage

This release adds support for Amazon S3 in the following regions:

• 20: AP_NORTHEAST_3
• 21: EU_NORTH_1
• 22: ME_SOUTH_1
• 23: US_GOV_EAST_1

This release adds maxVideoDuration to the start method to set the maximum duration of the MP4 slice file generated by the web page recording.

When the page status is a 404 or 5XX error, the page recording service automatically exits to avoid continuing to record the error page.

This release fixes an issue where audio was occasionally lost during a web page recording.

This version was released on June 10, 2021.

Network Geofencing

This release adds the region parameter to the acquire method so you can geofence your cloud recording service. If you specify a region, the cloud recording server connects only to the Agora servers within that region. The following regions are supported:

• Mainland China
• Asia, excluding Mainland China
• North America
• Europe

Third-party cloud storage

This release adds support for Microsoft Azure as a third-party storage solution.

This release optimizes the encoding settings for the web page recording, which reduce the bit rate without losing video quality.

Fixed security issues caused by vulnerabilities in Chrome V8.

• Adds the region parameter in the acquire method.
• Adds recorderStartTime in the web_recorder_started callback.
• Adds the web_recorder_reload callback.

This version was released on April 29, 2021.

New features

From this release, you can get MP4 files after a composite recording. Set avFileType as ["hls","mp4"] when you call the start method, and the recording service generates MP4 files once a recording starts.

For scenarios requiring high security, to guarantee the confidentiality, integrity, and authenticity of the data and to improve the computational efficiency of the data encryption, this release adds the following options in decryptionMode:

• 128-bit AES encryption, GCM mode.
• 256-bit AES encryption, GCM mode.

Once you enable the built-in encryption, all users in the same channel must use the same encryption mode and key.

Improvement

From this release, the recording service automatically detects and resolves the occasional audio loss issue during a web page recording, which causes the recorded page to refresh. If necessary, you need to guarantee that the status of your web application remains the same before and after a refresh.

API changes

• Adds enumerators 5 and 6 for decryptionMode in the start method.
• Adds the fileList field in the uploaded and backuped callbacks.

This version was released on March 31, 2021.

New feature

In composite recording mode, you can set the background image for both the video canvas and an individual user region. You can update the background image during a recording session. See Set the background color or background image for details.

Improvement

• Adds a retry logic for Kingsoft Cloud, in which the recording service retries if it fails to upload recorded files to Kingsoft Cloud.

Fixed issues

• Occasional file loss when uploading slice files to Tencent Cloud Object Storage (COS).
• In a web page recording, the recording service could not render emoji.
• When recording a static web page, the recording occasionally failed when the value of videoFps was too high.

API changes

• Adds the backgroundImage, defaultUserBackgroundImage and backgroundConfig fields in the transcodingConfig parameter of the start method.
• Adds the backgroundImage, defaultUserBackgroundImage and backgroundConfig fields in the clientRequest parameter of the updateLayout method.
• Adds the fileList field in the web_recorder_stopped callback.

This release raises the maximum resolution of video sources for web page recording. The recorded web page can now contain video that has a resolution of up to of 1920 × 1080.

This release includes the following new feature and API changes:

Adds web page recording mode. In this mode, the content and audio of a specified web page are recorded in a single file.

• Adds web as a value for mode in the request URL for web page recording.
• Adds scene as a new parameter in the acquire method for setting the application of the cloud recording resources.
• Adds "web_recorder_service" as an extension service in the extensionServiceConfig parameter of the start method.
• Adds the following callbacks, which report the status of web page recording:
• web_recorder_started
• web_recorder_stopped
• web_recorder_capability_limit

This release improved the stability and performance of the recording service, and fixed the following issues:

• Failure to subscribe to a low-quality stream.
• Occasional failure to record a whole call.

This release adds the following new feature and API changes:

New feature

In composite recording mode, you can now upload recorded files to ApsaraVideo for VoD for further processing.

API changes

• start adds the extensionServiceConfig parameter for specifying the configurations of ApsaraVideo for VoD.
• Adds the vod_started and the vod_triggered callbacks, which report the status of uploading recorded files to ApsaraVideo for VoD.

New features

This release provides a more flexible user ID subscription feature, enabling you to set up a whitelist or blacklist for audio and video subscriptions. You can also update the subscription lists during the cloud recording. For details, see Set up subscription lists.

This release adds the session_failover event, which occurs when the high availability mechanism is enabled and reports the new recording user ID.

API changes

• In start adds the unSubscribeVideoUids and unSubscribeAudioUids parameters for setting up subscription blacklists.
• Adds update for updating the subscription lists during the cloud recording.
• Adds the session_failover event.

This release adds support for Kingsoft Cloud.

Cloud Recording now supports taking screenshots of the video streams in individual recording mode. See Capture Screenshots for details.

This release improved the availability of the cloud recording service. When a server is disconnected or when the process is killed, the cloud recording automatically switches to a new server to resume the service. See Fault processing when a cloud recording server is disconnected or the process killed for more information.

API Changes

In the acquire method, adds the resourceExpiredHour parameter for setting the time limit for all cloud recording method calls.

Adds the [session_exit](../reference/rest-api-overview" >}}#a-name11a11-session_exit) event, which reports the exit status of the cloud recording service.

You can now use the query method to get the names of the recorded files immediately after the recording starts.

This release supports Tencent Cloud as one of the third-party cloud storages.

New features

The RESTful API adds individual recording mode, which supports recording the audio and video of each user ID separately. See Individual Recording for details. Meanwhile, Agora provides the Audio & Video File Merging script, which you can use to merge the audio and video files generated in individual mode. See Merge Audio and Video Files for details.

The RESTful API adds the subscribeAudioUids and subscribeVideoUids parameters, which allow you to record only specified user IDs.

The RESTful API adds the fileNamePrefix parameter, which allows you to specify the directory where you want to store the recorded files in the third-party cloud storage.

The RESTful API callback service adds the recorder_audio_stream_state_changed and recorder_video_stream_state_changed events to report the time when the state of the audio or video stream changes, as well as the corresponding user ID.

Agora provides the Format Converter script, which you can use to convert between multiple file formats, such as TS, MP3, and MP4. See Convert File Format for details.

You can achieve synchronized playback between the recorded files and other stream files, such as online whiteboards, courseware, and messages, by using the timestamp when the recording starts. You can get the timestamp by using the RESTful API callback service, or by parsing the M3U8 file. See Synchronized Playback for details.

Improvement

When an error occurs, you receive the error message in the HTTP response body, instead of just the error code. See Agora Cloud Recording RESTful API for detailed information about the error codes.

Fixed issue

When uploading fails after you use the wrong bucket and key values of the third-party cloud storage, Agora Cloud Recording returns an error instead of uploading the recorded files to Agora Cloud Backup.

New features

The RESTful API adds a customized layout for the recording video. See Set Video Layout for details.

You can set the mixedVideoLayout parameter as 3 and set the regions for each user in the layoutConfig parameter when starting a recording.

You can update the layout anytime during the recording by the updateLayout method.

The RESTful API adds the backgroundColor parameter to support customized background colors for the video layout.

To get the accurate starting time of a recording, the RESTful API provides the Unix timestamp of when the first slicing starts in the response of the query method. The RESTful API callback service adds the recorder_slice_start event to report the time when the first slicing starts and the time when the last recording fails.

Improvement

Optimizes the verification of whether resourceId corresponds with uid and cname when calling the RESTful API.

Fixed issue

Fixed minor issues in the default video layout (floating layout).

• Changes the default background color in the composite layout to black.
• Reduces video freeze under poor network conditions.

This release supports RESTful APIs. With the RESTful APIs, you can use Agora Cloud Recording through HTTP requests without integrating the SDK.

See the following documents for details:

• RESTful API Quickstart: Start cloud recording with RESTful APIs.
• RESTful API Reference: Details of the RESTful API methods.
• RESTful API Callback Service: Enable the callback service to receive notifications of Agora Cloud Recording.

This is the first release of Agora Cloud Recording with the following functions:

• High-quality voice and video recordings.
• Mixed-stream voice and video recordings of all users in a channel.
• Three composite video layouts: float (default), best fit, and vertical.
• Third-party cloud storage. Agora Cloud Recording supports Amazon S3, and Alibaba Cloud.
• Provides C++ and Java SDK packages.

• After you enable Notifications, your server receives the events that you subscribe to in the form of HTTPS requests.
• To improve communication security between the Notifications and your server, Agora SD-RTN™ uses signatures for identity verification.
• As of this release, you can use Notifications in conjunction with this product.