Control the retry policy from Rust
Using Golem's retry mechanism
Golem applies a retry mechanism to all workers. In case of a failure, Golem will automatically recover the worker to the point before the failure and retry the operation. An exponential backoff and an upper limit on the number of retries are applied.
If the maximum number of retries is reached, the worker will be marked as failed and no further invocations will be possible on it.
This mechanism is automatic and applied to all kind of failures. To rely on it, just let the Rust code panic.
Customizing the retry policy
The retry policy which controls the maximum number of retries and the exponential backoff is a global configuration of the Golem servers, but it can be customized for each worker.
For Rust we need to use the golem-rust
crate to do so.
First add it to the dependencies in your Cargo.toml
:
[dependencies]
golem-rust = { version = "0.3.0" }
golem-rust-macro = { version = "0.3.0" }
Then use the use_retry_policy
or with_retry_policy
functions to temporarily change the retry policy, either until the returned guard is dropped (the use_
variant) or for the provided closure (the with_
variant).
let _guard = use_retry_policy(RetryPolicy {
max_attempts: 10,
min_delay: 1000000000, // 1s
max_delay: 1000000000, // 1s
multiplier: 1,
});
// ...
// or
with_retry_policy(RetryPolicy {
max_attempts: 10,
min_delay: 1000000000, // 1s
max_delay: 1000000000, // 1s
multiplier: 1,
}, || {
// ...
});
The RetryPolicy
type itself is originated from Golem's WIT definition in the following way:
/// Configures how the executor retries failures
record retry-policy {
/// The maximum number of retries before the worker becomes permanently failed
max-attempts: u32,
/// The minimum delay between retries (applied to the first retry)
min-delay: duration,
/// The maximum delay between retries
max-delay: duration,
/// Multiplier applied to the delay on each retry to implement exponential backoff
multiplier: u32
}