Fault-tolerant execution#
By default, if a Trino node lacks the resources to execute a task or otherwise fails during query execution, the query fails and must be run again manually. The longer the runtime of a query, the more likely it is to be susceptible to such failures.
Fault-tolerant execution is a mechanism in Trino that enables a cluster to mitigate query failures by retrying queries or their component tasks in the event of failure. With fault-tolerant execution enabled, intermediate exchange data is spooled and can be re-used by another worker in the event of a worker outage or other fault during query execution.
Note
Fault tolerance does not apply to broken queries or other user error. For example, Trino does not spend resources retrying a query that fails because its SQL cannot be parsed.
For a step-by-step guide explaining how to configure a Trino cluster with fault-tolerant execution to improve query processing resilience, read Improve query processing resilience.
Configuration#
Fault-tolerant execution is disabled by default. To enable the feature, set the
retry-policy configuration property to either QUERY or TASK
depending on the desired retry policy.
retry-policy=QUERY
Warning
Setting retry-policy may cause queries to fail with connectors that do not
explicitly support fault-tolerant execution, resulting in a “This connector
does not support query retries” error message.
Support for fault-tolerant execution of SQL statements varies on a per-connector basis, with more details in the documentation for each connector. The following connectors support fault-tolerant execution:
The following configuration properties control the behavior of fault-tolerant execution on a Trino cluster:
Property name |
Description |
Default value |
|---|---|---|
|
Configures what is retried in the event of failure, either |
|
|
Data size of the coordinator’s in-memory buffer used by fault-tolerant execution to store output of query stages. If this buffer is filled during query execution, the query fails with a “Exchange manager must be configured for the failure recovery capabilities to be fully functional” error message unless an exchange manager is configured. |
|
|
Enable encryption of spooling data, see Encryption for details. Setting this property to false is not recommended if Trino processes sensitive data. |
|
Find further related properties in Properties reference, specifically in Resource management properties and Exchange properties.
Retry policy#
The retry-policy configuration property designates whether Trino retries
entire queries or a query’s individual tasks in the event of failure.
QUERY#
A QUERY retry policy instructs Trino to automatically retry a query in the
event of an error occuring on a worker node. A QUERY retry policy is
recommended when the majority of the Trino cluster’s workload consists of many
small queries.
By default Trino does not implement fault tolerance for queries whose result set
exceeds 32MB in size, such as SELECT statements that return a very
large data set to the user. This limit can be increased by modifying the
exchange.deduplication-buffer-size configuration property to be greater than
the default value of 32MB, but this results in higher memory usage on the
coordinator.
To enable fault-tolerant execution on queries with a larger result set, it is strongly recommended to configure an exchange manager that utilizes external storage for spooled data and therefore allows for storage of spilled data beyond the in-memory buffer size.
TASK#
A TASK retry policy instructs Trino to retry individual query tasks in the event of failure. You must configure an
exchange manager to use the task retry policy.
This policy is recommended when executing large batch queries, as the cluster
can more efficiently retry smaller tasks within the query rather than retry the
whole query.
When a cluster is configured with a TASK retry policy, some relevant
configuration properties have their default values changed to follow best
practices for a fault-tolerant cluster. However, this automatic change does not
affect clusters that have these properties manually configured. If you have
any of the following properties configured in the config.properties file on
a cluster with a TASK retry policy, it is strongly recommended to set the
task.low-memory-killer.policy
query management property to
total-reservation-on-blocked-nodes, or queries may need to be manually killed
if the cluster runs out of memory.
Note
A TASK retry policy is best suited for large batch queries, but this
policy can result in higher latency for short-running queries executed in high
volume. As a best practice, it is recommended to run a dedicated cluster
with a TASK retry policy for large batch queries, separate from another
cluster that handles short queries.
Encryption#
Trino encrypts data before spooling it to storage. This prevents access to query data by anyone besides the Trino cluster that wrote it, including administrators of the storage system. A new encryption key is randomly generated for every query, and the key is discarded once a query is completed.
Advanced configuration#
You can further configure fault-tolerant execution with the following configuration properties. The default values for these properties should work for most deployments, but you can change these values for testing or troubleshooting purposes.
Retry limits#
The following configuration properties control the thresholds at which queries/tasks are no longer retried in the event of repeated failures:
Property name |
Description |
Default value |
Retry policy |
|---|---|---|---|
|
Maximum number of times Trino may attempt to retry a query before declaring the query as failed. |
|
Only |
|
Maximum number of times Trino may attempt to retry a single task before declaring the query as failed. |
|
Only |
|
Minimum time that a failed query or task must wait
before it is retried. May be overridden with the |
|
|
|
Maximum time that a failed query or task must
wait before it is retried. Wait time is increased on each subsequent
failure. May be overridden with the |
|
|
|
Factor by which retry delay is increased on each query or task failure. May
be overridden with the |
|
|
Task sizing#
With a TASK retry policy, it is important to manage the amount of data
processed in each task. If tasks are too small, the management of task
coordination can take more processing time and resources than executing the task
itself. If tasks are too large, then a single task may require more resources
than are available on any one node and therefore prevent the query from
completing.
Trino supports limited automatic task sizing. If issues are occurring
during fault-tolerant task execution, you can configure the following
configuration properties to manually control task sizing. These configuration
properties only apply to a TASK retry policy.
Property name |
Description |
Default value |
|---|---|---|
|
Standard split data size processed by tasks that read data from source tables. Value is interpreted with split weight taken into account. If the weight of splits produced by a catalog denotes that they are lighter or heavier than “standard” split, then the number of splits processed by a single task is adjusted accordingly. May be overridden for the current session with the
|
|
|
Maximum number of splits processed by a single task. This value is not split weight-adjusted and serves as protection against situations where catalogs report an incorrect split weight. May be overridden for the current session with the
|
|
|
The number of tasks created for any given non-writer stage of arbitrary distribution before task size is increased. |
|
|
Growth factor for adaptive sizing of non-writer tasks of arbitrary distribution for fault-tolerant execution. Lower bound is 1.0. For every task size increase, new task target size is old task target size multiplied by this growth factor. |
|
|
Initial/minimum target input data size for non-writer tasks of arbitrary distribution of fault-tolerant execution. |
|
|
Maximum target input data size for each non-writer task of arbitrary distribution of fault-tolerant execution. |
|
|
The number of tasks created for any given writer stage of arbitrary distribution before task size is increased. |
|
|
Growth factor for adaptive sizing of writer tasks of arbitrary distribution for fault-tolerant execution. Lower bound is 1.0. For every task size increase, new task target size is old task target size multiplied by this growth factor. |
|
|
Initial/minimum target input data size for writer tasks of arbitrary distribution of fault-tolerant execution. |
|
|
Maximum target input data size for writer tasks of arbitrary distribution of fault-tolerant execution. |
|
|
Target input data size for non-writer tasks of hash distribution of fault-tolerant execution. |
|
|
Target input data size of writer tasks of hash distribution of fault-tolerant execution. |
|
|
Soft upper bound on number of writer tasks in a stage of hash distribution of fault-tolerant execution. |
|
Node allocation#
With a TASK retry policy, nodes are allocated to tasks based on available
memory and estimated memory usage. If task failure occurs due to exceeding
available memory on a node, the task is restarted with a request to allocate the
full node for its execution.
The initial task memory-requirements estimation is static and configured with
the fault-tolerant-execution-task-memory configuration property. This property only
applies to a TASK retry policy.
Property name |
Description |
Default value |
|---|---|---|
|
Initial task memory data size estimation
used for bin-packing when allocating nodes for tasks. May be overridden
for the current session with the
|
|
Other tuning#
The following additional configuration property can be used to manage fault-tolerant execution:
Property name |
Description |
Default value |
Retry policy |
|---|---|---|---|
|
Maximum data size of memory to be used to store task descriptors for fault tolerant queries on coordinator. Extra memory is needed to be able to reschedule tasks in case of a failure. |
(JVM heap size * 0.15) |
Only |
|
Maximum number of partitions to use for distributed joins and aggregations,
similar in function to the |
|
Only |
|
Minimum number of partitions to use for distributed joins and aggregations,
similar in function to the |
|
Only |
|
Minimum number of partitions to use for distributed joins and aggregations
in write queries, similar in function to the
|
|
Only |
|
Allow for up to configured number of tasks to wait for node allocation per query, before pausing scheduling for other tasks from this query. |
|
Only |
Exchange manager#
Exchange spooling is responsible for storing and managing spooled data for fault-tolerant execution. You can configure a filesystem-based exchange manager that stores spooled data in a specified location, such as AWS S3 and S3-compatible systems, Azure Blob Storage, Google Cloud Storage, or HDFS.
Configuration#
To configure an exchange manager, create a new
etc/exchange-manager.properties configuration file on the coordinator and
all worker nodes. In this file, set the exchange-manager.name configuration
property to filesystem or hdfs, and set additional configuration properties as needed
for your storage solution.
The following table lists the available configuration properties for
exchange-manager.properties, their default values, and which filesystem(s)
the property may be configured for:
Property name |
Description |
Default value |
Supported filesystem |
|---|---|---|---|
|
Comma-separated list of URI locations that the exchange manager uses to store spooling data. |
Any |
|
|
The minimum buffer pool size for an exchange sink. The larger the buffer pool size, the larger the write parallelism and memory usage. |
|
Any |
|
The number of buffers per partition in the buffer pool. The larger the buffer pool size, the larger the write parallelism and memory usage. |
|
Any |
|
Max data size of files written by exchange sinks. |
|
Any |
|
Number of concurrent readers to read from spooling storage. The larger the number of concurrent readers, the larger the read parallelism and memory usage. |
|
Any |
|
AWS access key to use. Required for a connection to AWS S3 and GCS, can be ignored for other S3 storage systems. |
AWS S3, GCS |
|
|
AWS secret key to use. Required for a connection to AWS S3 and GCS, can be ignored for other S3 storage systems. |
AWS S3, GCS |
|
|
IAM role to assume. |
AWS S3, GCS |
|
|
External ID for the IAM role trust policy. |
AWS S3, GCS |
|
|
Region of the S3 bucket. |
AWS S3, GCS |
|
|
S3 storage endpoint server if using an S3-compatible storage system that
is not AWS. If using AWS S3, this can be ignored unless HTTPS is required
by an AWS bucket policy. If TLS is required, then this property can be
set to an https endpoint such as |
Any S3-compatible storage |
|
|
Maximum number of times the exchange manager’s S3 client should retry a request. |
|
Any S3-compatible storage |
|
Enables using path-style access for all requests to S3. |
|
Any S3-compatible storage |
|
Part data size for S3 multi-part upload. |
|
Any S3-compatible storage |
|
Path to the JSON file that contains your Google Cloud Platform service
account key. Not to be set together with |
GCS |
|
|
Your Google Cloud Platform service account key in JSON format. Not to be set
together with |
GCS |
|
|
Azure blob endpoint used to access the spooling container. Not to be set
together with |
Azure Blob Storage |
|
|
Connection string used to access the spooling container. Not to be set
together with |
Azure Blob Storage |
|
|
Block data size for Azure block blob parallel upload. |
|
Azure Blob Storage |
|
Maximum number of times the exchange manager’s Azure client should retry a request. |
|
Azure Blob Storage |
|
Block data size for HDFS storage. |
|
HDFS |
|
Comma-separated list of paths to HDFS configuration files, for example
|
HDFS |
To reduce the exchange manager’s overall I/O load, the
exchange.compression-codec configuration property defaults to LZ4. In
addition, File compression and decompression is automatically performed and some details can
be configured.
It is also recommended to configure a bucket lifecycle rule to automatically expire abandoned objects in the event of a node crash.
AWS S3#
The following example exchange-manager.properties configuration specifies an
AWS S3 bucket as the spooling storage destination. Note that the destination
does not have to be in AWS, but can be any S3-compatible storage system. While
the exchange manager is designed to support S3-compatible storage systems, only
AWS S3 and MinIO are tested for compatibility. For other storage systems,
perform your own testing and consult your vendor for more information.
exchange-manager.name=filesystem
exchange.base-directories=s3://exchange-spooling-bucket
exchange.s3.region=us-west-1
exchange.s3.aws-access-key=example-access-key
exchange.s3.aws-secret-key=example-secret-key
You can configure multiple S3 buckets for the exchange manager to distribute
spooled data across buckets, reducing the I/O load on any one bucket. If a query
fails with the error message
“software.amazon.awssdk.services.s3.model.S3Exception: Please reduce your
request rate”, this indicates that the workload is I/O intensive, and you should
specify multiple S3 buckets in exchange.base-directories to balance the
load:
exchange.base-directories=s3://exchange-spooling-bucket-1,s3://exchange-spooling-bucket-2
Azure Blob Storage#
The following example exchange-manager.properties configuration specifies an
Azure Blob Storage container as the spooling storage destination. You must use
Azure Blob Storage, not Azure Data Lake Storage or any other hierarchical
storage option in Azure.
exchange-manager.name=filesystem
exchange.base-directories=abfs://container_name@account_name.dfs.core.windows.net
exchange.azure.connection-string=connection-string
Google Cloud Storage#
To enable exchange spooling on GCS in Trino, change the request endpoint to the
https://storage.googleapis.com Google storage URI, and configure your AWS
access/secret keys to use the GCS HMAC keys. If you deploy Trino on GCP, you
must either create a service account with access to your spooling bucket or
configure the key path to your GCS credential file.
For more information on GCS’s S3 compatibility, refer to the Google Cloud documentation on S3 migration.
The following example exchange-manager.properties configuration specifies a
GCS bucket as the spooling storage destination.
exchange-manager.name=filesystem
exchange.base-directories=gs://exchange-spooling-bucket
exchange.s3.region=us-west-1
exchange.s3.aws-access-key=example-access-key
exchange.s3.aws-secret-key=example-secret-key
exchange.s3.endpoint=https://storage.googleapis.com
exchange.gcs.json-key-file-path=/path/to/gcs_keyfile.json
HDFS#
The following exchange-manager.properties configuration example specifies HDFS
as the spooling storage destination.
exchange-manager.name=hdfs
exchange.base-directories=hadoop-master:9000/exchange-spooling-directory
hdfs.config.resources=/usr/lib/hadoop/etc/hadoop/core-site.xml
Local filesystem storage#
The following example exchange-manager.properties configuration specifies a
local directory, /tmp/trino-exchange-manager, as the spooling storage
destination.
Note
It is only recommended to use a local filesystem for exchange in standalone, non-production clusters. A local directory can only be used for exchange in a distributed cluster if the exchange directory is shared and accessible from all worker nodes.
exchange-manager.name=filesystem
exchange.base-directories=/tmp/trino-exchange-manager
Adaptive plan optimizations#
Fault-tolerant execution mode offers several adaptive plan optimizations that adjust query execution plans dynamically based on runtime statistics. For more information, see Adaptive plan optimizations.