FirehoseClient

Deletes a Firehose stream and its data.

You can delete a Firehose stream only if it is in one of the following states: ACTIVE, DELETING, CREATING_FAILED, or DELETING_FAILED. You can't delete a Firehose stream that is in the CREATING state. To check the state of a Firehose stream, use DescribeDeliveryStream.

DeleteDeliveryStream is an asynchronous API. When an API request to DeleteDeliveryStream succeeds, the Firehose stream is marked for deletion, and it goes into the DELETING state.While the Firehose stream is in the DELETING state, the service might continue to accept records, but it doesn't make any guarantees with respect to delivering the data. Therefore, as a best practice, first stop any applications that are sending records before you delete a Firehose stream.

Removal of a Firehose stream that is in the DELETING state is a low priority operation for the service. A stream may remain in the DELETING state for several minutes. Therefore, as a best practice, applications should not wait for streams in the DELETING state to be removed.

Describes the specified Firehose stream and its status. For example, after your Firehose stream is created, call DescribeDeliveryStream to see whether the Firehose stream is ACTIVE and therefore ready for data to be sent to it.

If the status of a Firehose stream is CREATING_FAILED, this status doesn't change, and you can't invoke CreateDeliveryStream again on it. However, you can invoke the DeleteDeliveryStream operation to delete it. If the status is DELETING_FAILED, you can force deletion by invoking DeleteDeliveryStream again but with DeleteDeliveryStreamInput$AllowForceDelete set to true.

Lists your Firehose streams in alphabetical order of their names.

The number of Firehose streams might be too large to return using a single call to ListDeliveryStreams. You can limit the number of Firehose streams returned, using the Limit parameter. To determine whether there are more delivery streams to list, check the value of HasMoreDeliveryStreams in the output. If there are more Firehose streams to list, you can request them by calling this operation again and setting the ExclusiveStartDeliveryStreamName parameter to the name of the last Firehose stream returned in the last call.

Lists the tags for the specified Firehose stream. This operation has a limit of five transactions per second per account.

Writes multiple data records into a Firehose stream in a single call, which can achieve higher throughput per producer than when writing single records. To write single data records into a Firehose stream, use PutRecord. Applications using these operations are referred to as producers.

Firehose accumulates and publishes a particular metric for a customer account in one minute intervals. It is possible that the bursts of incoming bytes/records ingested to a Firehose stream last only for a few seconds. Due to this, the actual spikes in the traffic might not be fully visible in the customer's 1 minute CloudWatch metrics.

For information about service quota, see Amazon Firehose Quota .

Each PutRecordBatch request supports up to 500 records. Each record in the request can be as large as 1,000 KB (before base64 encoding), up to a limit of 4 MB for the entire request. These limits cannot be changed.

You must specify the name of the Firehose stream and the data record when using PutRecord. The data record consists of a data blob that can be up to 1,000 KB in size, and any kind of data. For example, it could be a segment from a log file, geographic location data, website clickstream data, and so on.

For multi record de-aggregation, you can not put more than 500 records even if the data blob length is less than 1000 KiB. If you include more than 500 records, the request succeeds but the record de-aggregation doesn't work as expected and transformation lambda is invoked with the complete base64 encoded data blob instead of de-aggregated base64 decoded records.

Firehose buffers records before delivering them to the destination. To disambiguate the data blobs at the destination, a common solution is to use delimiters in the data, such as a newline (n) or some other character unique within the data. This allows the consumer application to parse individual data items when reading the data from the destination.

The PutRecordBatch response includes a count of failed records, FailedPutCount, and an array of responses, RequestResponses. Even if the PutRecordBatch call succeeds, the value of FailedPutCount may be greater than 0, indicating that there are records for which the operation didn't succeed. Each entry in the RequestResponses array provides additional information about the processed record. It directly correlates with a record in the request array using the same ordering, from the top to the bottom. The response array always includes the same number of records as the request array. RequestResponses includes both successfully and unsuccessfully processed records. Firehose tries to process all records in each PutRecordBatch request. A single record failure does not stop the processing of subsequent records.

A successfully processed record includes a RecordId value, which is unique for the record. An unsuccessfully processed record includes ErrorCode and ErrorMessage values. ErrorCode reflects the type of error, and is one of the following values: ServiceUnavailableException or InternalFailure. ErrorMessage provides more detailed information about the error.

If there is an internal server error or a timeout, the write might have completed or it might have failed. If FailedPutCount is greater than 0, retry the request, resending only those records that might have failed processing. This minimizes the possible duplicate records and also reduces the total bytes sent (and corresponding charges). We recommend that you handle any duplicates at the destination.

If PutRecordBatch throws ServiceUnavailableException, the API is automatically reinvoked (retried) 3 times. If the exception persists, it is possible that the throughput limits have been exceeded for the Firehose stream.

Re-invoking the Put API operations (for example, PutRecord and PutRecordBatch) can result in data duplicates. For larger data assets, allow for a longer time out before retrying Put API operations.

Data records sent to Firehose are stored for 24 hours from the time they are added to a Firehose stream as it attempts to send the records to the destination. If the destination is unreachable for more than 24 hours, the data is no longer available.

Writes a single data record into an Firehose stream. To write multiple data records into a Firehose stream, use PutRecordBatch. Applications using these operations are referred to as producers.

By default, each Firehose stream can take in up to 2,000 transactions per second, 5,000 records per second, or 5 MB per second. If you use PutRecord and PutRecordBatch, the limits are an aggregate across these two operations for each Firehose stream. For more information about limits and how to request an increase, see Amazon Firehose Limits .

Firehose accumulates and publishes a particular metric for a customer account in one minute intervals. It is possible that the bursts of incoming bytes/records ingested to a Firehose stream last only for a few seconds. Due to this, the actual spikes in the traffic might not be fully visible in the customer's 1 minute CloudWatch metrics.

You must specify the name of the Firehose stream and the data record when using PutRecord. The data record consists of a data blob that can be up to 1,000 KiB in size, and any kind of data. For example, it can be a segment from a log file, geographic location data, website clickstream data, and so on.

For multi record de-aggregation, you can not put more than 500 records even if the data blob length is less than 1000 KiB. If you include more than 500 records, the request succeeds but the record de-aggregation doesn't work as expected and transformation lambda is invoked with the complete base64 encoded data blob instead of de-aggregated base64 decoded records.

Firehose buffers records before delivering them to the destination. To disambiguate the data blobs at the destination, a common solution is to use delimiters in the data, such as a newline (n) or some other character unique within the data. This allows the consumer application to parse individual data items when reading the data from the destination.

The PutRecord operation returns a RecordId, which is a unique string assigned to each record. Producer applications can use this ID for purposes such as auditability and investigation.

If the PutRecord operation throws a ServiceUnavailableException, the API is automatically reinvoked (retried) 3 times. If the exception persists, it is possible that the throughput limits have been exceeded for the Firehose stream.

Re-invoking the Put API operations (for example, PutRecord and PutRecordBatch) can result in data duplicates. For larger data assets, allow for a longer time out before retrying Put API operations.

Data records sent to Firehose are stored for 24 hours from the time they are added to a Firehose stream as it tries to send the records to the destination. If the destination is unreachable for more than 24 hours, the data is no longer available.

Enables server-side encryption (SSE) for the Firehose stream.

This operation is asynchronous. It returns immediately. When you invoke it, Firehose first sets the encryption status of the stream to ENABLING, and then to ENABLED. The encryption status of a Firehose stream is the Status property in DeliveryStreamEncryptionConfiguration. If the operation fails, the encryption status changes to ENABLING_FAILED. You can continue to read and write data to your Firehose stream while the encryption status is ENABLING, but the data is not encrypted. It can take up to 5 seconds after the encryption status changes to ENABLED before all records written to the Firehose stream are encrypted. To find out whether a record or a batch of records was encrypted, check the response elements PutRecordOutput$Encrypted and PutRecordBatchOutput$Encrypted, respectively.

To check the encryption status of a Firehose stream, use DescribeDeliveryStream.

Even if encryption is currently enabled for a Firehose stream, you can still invoke this operation on it to change the ARN of the CMK or both its type and ARN. If you invoke this method to change the CMK, and the old CMK is of type CUSTOMER_MANAGED_CMK, Firehose schedules the grant it had on the old CMK for retirement. If the new CMK is of type CUSTOMER_MANAGED_CMK, Firehose creates a grant that enables it to use the new CMK to encrypt and decrypt data and to manage the grant.

For the KMS grant creation to be successful, the Firehose API operations StartDeliveryStreamEncryption and CreateDeliveryStream should not be called with session credentials that are more than 6 hours old.

If a Firehose stream already has encryption enabled and then you invoke this operation to change the ARN of the CMK or both its type and ARN and you get ENABLING_FAILED, this only means that the attempt to change the CMK failed. In this case, encryption remains enabled with the old CMK.

If the encryption status of your Firehose stream is ENABLING_FAILED, you can invoke this operation again with a valid CMK. The CMK must be enabled and the key policy mustn't explicitly deny the permission for Firehose to invoke KMS encrypt and decrypt operations.

You can enable SSE for a Firehose stream only if it's a Firehose stream that uses DirectPut as its source.

The StartDeliveryStreamEncryption and StopDeliveryStreamEncryption operations have a combined limit of 25 calls per Firehose stream per 24 hours. For example, you reach the limit if you call StartDeliveryStreamEncryption 13 times and StopDeliveryStreamEncryption 12 times for the same Firehose stream in a 24-hour period.

Disables server-side encryption (SSE) for the Firehose stream.

This operation is asynchronous. It returns immediately. When you invoke it, Firehose first sets the encryption status of the stream to DISABLING, and then to DISABLED. You can continue to read and write data to your stream while its status is DISABLING. It can take up to 5 seconds after the encryption status changes to DISABLED before all records written to the Firehose stream are no longer subject to encryption. To find out whether a record or a batch of records was encrypted, check the response elements PutRecordOutput$Encrypted and PutRecordBatchOutput$Encrypted, respectively.

To check the encryption state of a Firehose stream, use DescribeDeliveryStream.

If SSE is enabled using a customer managed CMK and then you invoke StopDeliveryStreamEncryption, Firehose schedules the related KMS grant for retirement and then retires it after it ensures that it is finished delivering records to the destination.

The StartDeliveryStreamEncryption and StopDeliveryStreamEncryption operations have a combined limit of 25 calls per Firehose stream per 24 hours. For example, you reach the limit if you call StartDeliveryStreamEncryption 13 times and StopDeliveryStreamEncryption 12 times for the same Firehose stream in a 24-hour period.

Adds or updates tags for the specified Firehose stream. A tag is a key-value pair that you can define and assign to Amazon Web Services resources. If you specify a tag that already exists, the tag value is replaced with the value that you specify in the request. Tags are metadata. For example, you can add friendly names and descriptions or other types of information that can help you distinguish the Firehose stream. For more information about tags, see Using Cost Allocation Tags  in the Amazon Web Services Billing and Cost Management User Guide.

Each Firehose stream can have up to 50 tags.

This operation has a limit of five transactions per second per account.

Removes tags from the specified Firehose stream. Removed tags are deleted, and you can't recover them after this operation successfully completes.

If you specify a tag that doesn't exist, the operation ignores it.

This operation has a limit of five transactions per second per account.