/build/source/nativelink-store/src/s3_store.rs

Source
// Copyright 2024 The NativeLink Authors. All rights reserved.
//
// Licensed under the Functional Source License, Version 1.1, Apache 2.0 Future License (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//    See LICENSE file for details
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

use core::cmp;
use core::pin::Pin;
use core::time::Duration;
use std::borrow::Cow;
use std::sync::Arc;

use async_trait::async_trait;
use aws_config::default_provider::credentials;
use aws_config::provider_config::ProviderConfig;
use aws_config::{AppName, BehaviorVersion};
use aws_sdk_s3::Client;
use aws_sdk_s3::config::Region;
use aws_sdk_s3::operation::create_multipart_upload::CreateMultipartUploadOutput;
use aws_sdk_s3::operation::get_object::GetObjectError;
use aws_sdk_s3::operation::head_object::HeadObjectError;
use aws_sdk_s3::primitives::ByteStream; // SdkBody
use aws_sdk_s3::types::builders::{CompletedMultipartUploadBuilder, CompletedPartBuilder};
use aws_smithy_types::body::SdkBody;
use futures::future::FusedFuture;
use futures::stream::{FuturesUnordered, unfold};
use futures::{FutureExt, StreamExt, TryFutureExt, TryStreamExt};
use nativelink_config::stores::ExperimentalAwsSpec;
// Note: S3 store should be very careful about the error codes it returns
// when in a retryable wrapper. Always prefer Code::Aborted or another
// retryable code over Code::InvalidArgument or make_input_err!().
// ie: Don't import make_input_err!() to help prevent this.
use nativelink_error::{Code, Error, ResultExt, make_err};
use nativelink_metric::MetricsComponent;
use nativelink_util::buf_channel::{
    DropCloserReadHalf, DropCloserWriteHalf, make_buf_channel_pair,
};
use nativelink_util::health_utils::{HealthRegistryBuilder, HealthStatus, HealthStatusIndicator};
use nativelink_util::instant_wrapper::InstantWrapper;
use nativelink_util::retry::{Retrier, RetryResult};
use nativelink_util::store_trait::{
    RemoveCallback, StoreDriver, StoreKey, StoreOptimizations, UploadSizeInfo,
};
use parking_lot::Mutex;
use tokio::sync::mpsc;
use tokio::time::sleep;
use tracing::{error, info};

use crate::cas_utils::is_zero_digest;
use crate::common_s3_utils::{BodyWrapper, TlsClient};

// S3 object cannot be larger than this number. See:
// https://docs.aws.amazon.com/AmazonS3/latest/userguide/qfacts.html
const MAX_UPLOAD_SIZE: u64 = 48 * 1024 * 1024 * 1024 * 1024; // 48TiB (technically should be 48.8 TiB, but close enough)

// S3 parts cannot be smaller than this number. See:
// https://docs.aws.amazon.com/AmazonS3/latest/userguide/qfacts.html
const MIN_MULTIPART_SIZE: u64 = 5 * 1024 * 1024; // 5MB.

// S3 parts cannot be larger than this number. See:
// https://docs.aws.amazon.com/AmazonS3/latest/userguide/qfacts.html
const MAX_MULTIPART_SIZE: u64 = 5 * 1024 * 1024 * 1024; // 5GB.

// S3 parts cannot be more than this number. See:
// https://docs.aws.amazon.com/AmazonS3/latest/userguide/qfacts.html
// Note: Type 'u64' chosen to simplify calculations
const MAX_UPLOAD_PARTS: u64 = 10_000;

// Default max buffer size for retrying upload requests.
// Note: If you change this, adjust the docs in the config.
const DEFAULT_MAX_RETRY_BUFFER_PER_REQUEST: usize = 5 * 1024 * 1024; // 5MB.

// Default limit for concurrent part uploads per multipart upload.
// Note: If you change this, adjust the docs in the config.
const DEFAULT_MULTIPART_MAX_CONCURRENT_UPLOADS: usize = 10;

#[derive(Debug, MetricsComponent)]
pub struct S3Store<NowFn> {
    s3_client: Arc<Client>,
    now_fn: NowFn,
    #[metric(help = "The bucket name for the S3 store")]
    bucket: String,
    #[metric(help = "The key prefix for the S3 store")]
    key_prefix: String,
    retrier: Retrier,
    #[metric(help = "The number of seconds to consider an object expired")]
    consider_expired_after_s: i64,
    #[metric(help = "The number of bytes to buffer for retrying requests")]
    max_retry_buffer_per_request: usize,
    #[metric(help = "The number of concurrent uploads allowed for multipart uploads")]
    multipart_max_concurrent_uploads: usize,

    remove_callbacks: Mutex<Vec<RemoveCallback>>,
}

impl<I, NowFn> S3Store<NowFn>
where
    I: InstantWrapper,
    NowFn: Fn() -> I + Send + Sync + Unpin + 'static,
{
    pub async fn new(spec: &ExperimentalAwsSpec, now_fn: NowFn) -> Result<Arc<Self>, Error> {
        let jitter_fn = spec.common.retry.make_jitter_fn();
        let s3_client = {
            let http_client = TlsClient::new(&spec.common.clone());

            let credential_provider = credentials::DefaultCredentialsChain::builder()
                .configure(
                    ProviderConfig::without_region()
                        .with_region(Some(Region::new(Cow::Owned(spec.region.clone()))))
                        .with_http_client(http_client.clone()),
                )
                .build()
                .await;

            let config = aws_config::defaults(BehaviorVersion::latest())
                .credentials_provider(credential_provider)
                .app_name(AppName::new("nativelink").expect("valid app name"))
                .timeout_config(
                    aws_config::timeout::TimeoutConfig::builder()
                        .connect_timeout(Duration::from_secs(15))
                        .build(),
                )
                .region(Region::new(Cow::Owned(spec.region.clone())))
                .http_client(http_client)
                .load()
                .await;

            Client::new(&config)
        };
        Self::new_with_client_and_jitter(spec, s3_client, jitter_fn, now_fn)
    }

    pub fn new_with_client_and_jitter(
        spec: &ExperimentalAwsSpec,
        s3_client: Client,
        jitter_fn: Arc<dyn Fn(Duration) -> Duration + Send + Sync>,
        now_fn: NowFn,
    ) -> Result<Arc<Self>, Error> {
        Ok(Arc::new(Self {
            s3_client: Arc::new(s3_client),
            now_fn,
            bucket: spec.bucket.clone(),
            key_prefix: spec
                .common
                .key_prefix
                .as_ref()
                .unwrap_or(&String::new())
                .clone(),
            retrier: Retrier::new(
                Arc::new(|duration| Box::pin2(sleep2(duration2))),
                jitter_fn,
                spec.common.retry.clone(),
            ),
            consider_expired_after_s: i64::from(spec.common.consider_expired_after_s),
            max_retry_buffer_per_request: spec
                .common
                .max_retry_buffer_per_request
                .unwrap_or(DEFAULT_MAX_RETRY_BUFFER_PER_REQUEST),
            multipart_max_concurrent_uploads: spec
                .common
                .multipart_max_concurrent_uploads
                .map_or(DEFAULT_MULTIPART_MAX_CONCURRENT_UPLOADS, |v| v),
            remove_callbacks: Mutex::new(Vec::new()),
        }))
    }

    fn make_s3_path(&self, key: &StoreKey<'_>) -> String {
        format!("{}{}", self.key_prefix, key.as_str())
    }

    async fn has(self: Pin<&Self>, digest: StoreKey<'_>) -> Result<Option<u64>, Error> {
        let digest_clone = digest.into_owned();
        self.retrier
            .retry7(unfold7(()7, move |state| {
                let local_digest = digest_clone.clone();
                async move {
                    let result = self
                        .s3_client
                        .head_object()
                        .bucket(&self.bucket)
                        .key(self.make_s3_path(&local_digest))
                        .send()
                        .await;

                    match result {
                        Ok(head_object_output) => {
                            if self.consider_expired_after_s != 0
                                && let Some(last_modified) = head_object_output.last_modified
                            {
                                let now_s = (self.now_fn)().unix_timestamp() as i64;
                                if last_modified.secs() + self.consider_expired_after_s <= now_s {
                                    let remove_callbacks = self.remove_callbacks.lock().clone();
                                    let mut callbacks: FuturesUnordered<_> = remove_callbacks
                                        .iter()
                                        .map(|callback| callback0.callback0(local_digest0.borrow0()))
                                        .collect();
                                    while callbacks.next().await.is_some() {}0
                                    return Some((RetryResult::Ok(None), state));
                                }
                            }
                            let Some(length) = head_object_output.content_length else {
                                return Some((RetryResult::Ok(None), state));
                            };
                            if length >= 0 {
                                return Some((RetryResult::Ok(Some(length as u64)), state));
                            }
                            Some((
                                RetryResult::Err(make_err!(
                                    Code::InvalidArgument,
                                    "Negative content length in S3: {length:?}",
                                )),
                                state,
                            ))
                        }
                        Err(sdk_error) => match sdk_error.into_service_error() {
                            HeadObjectError::NotFound(_) => Some((RetryResult::Ok(None), state)),
                            other => Some((
                                RetryResult::Retry(
                                    Error::from_std_err(Code::Unavailable, &other)
                                        .append("Unhandled HeadObjectError in S3"),
                                ),
                                state,
                            )),
                        },
                    }
                }
            }))
            .await
    }
}

#[async_trait]
impl<I, NowFn> StoreDriver for S3Store<NowFn>
where
    I: InstantWrapper,
    NowFn: Fn() -> I + Send + Sync + Unpin + 'static,
{
    async fn post_init(self: Arc<Self>) -> Result<(), Error> {
        Ok(())
    }

    async fn has_with_results(
        self: Pin<&Self>,
        keys: &[StoreKey<'_>],
        results: &mut [Option<u64>],
    ) -> Result<(), Error> {
        keys.iter()
            .zip(results.iter_mut())
            .map(|(key, result)| async move {
                // We need to do a special pass to ensure our zero key exist.
                if is_zero_digest(key.borrow()) {
                    *result = Some(0);
                    return Ok::<_, Error>(());
                }
                *result = self.has(key.borrow()).await?0;
                Ok::<_, Error>(())
            })
            .collect::<FuturesUnordered<_>>()
            .try_collect()
            .await
    }

    fn optimized_for(&self, optimization: StoreOptimizations) -> bool {
        matches!(optimization, StoreOptimizations::LazyExistenceOnSync)
    }

    async fn update(
        self: Pin<&Self>,
        digest: StoreKey<'_>,
        mut reader: DropCloserReadHalf,
        upload_size: UploadSizeInfo,
    ) -> Result<u64, Error> {
        let s3_path = &self.make_s3_path(&digest);

        let max_size = match upload_size {
            UploadSizeInfo::ExactSize(sz) | UploadSizeInfo::MaxSize(sz) => sz,
        };

        // Sanity check S3 maximum upload size.
        if max_size > MAX_UPLOAD_SIZE {
            return Err(make_err!(
                Code::FailedPrecondition,
                "File size exceeds max of {MAX_UPLOAD_SIZE}"
            ));
        }

        // Note(aaronmondal) It might be more optimal to use a different
        // heuristic here, but for simplicity we use a hard coded value.
        // Anything going down this if-statement will have the advantage of only
        // 1 network request for the upload instead of minimum of 3 required for
        // multipart upload requests.
        //
        // Note(aaronmondal) If the upload size is not known, we go down the multipart upload path.
        // This is not very efficient, but it greatly reduces the complexity of the code.
        if max_size < MIN_MULTIPART_SIZE && matches!(upload_size, UploadSizeInfo::ExactSize(_)) {
            let UploadSizeInfo::ExactSize(sz) = upload_size else {
                unreachable!("upload_size must be UploadSizeInfo::ExactSize here");
            };
            reader.set_max_recent_data_size(
                u64::try_from(self.max_retry_buffer_per_request)
                    .err_tip(|| "Could not convert max_retry_buffer_per_request to u64")?,
            );
            return self
                .retrier
                .retry(unfold(reader, move |mut reader| async move {
                    // We need to make a new pair here because the aws sdk does not give us
                    // back the body after we send it in order to retry.
                    let (mut tx, rx) = make_buf_channel_pair();

                    // Upload the data to the S3 backend.
                    let result = {
                        let reader_ref = &mut reader;
                        let (upload_res, bind_res) = tokio::join!(
                            self.s3_client
                                .put_object()
                                .bucket(&self.bucket)
                                .key(s3_path.clone())
                                .content_length(sz as i64)
                                .body(ByteStream::from_body_1_x(BodyWrapper {
                                    reader: rx,
                                    size: sz,
                                }))
                                .send()
                                .map_ok_or_else(|e| Err(Error::from_std_err0(Code::Aborted0, &e0)), |_| Ok(sz)),
                            // Stream all data from the reader channel to the writer channel.
                            tx.bind_buffered(reader_ref)
                        );
                        match (upload_res, bind_res) {
                            (Ok(size), Ok(())) => Ok(size),
                            (Err(e), _) | (_, Err(e)) => Err(e),
                        }
                        .err_tip(|| "Failed to upload file to s3 in single chunk")
                    };

                    // If we failed to upload the file, check to see if we can retry.
                    let retry_result = result.map_or_else(|mut err| {0
                        // Ensure our code is Code::Aborted, so the client can retry if possible.
                        err.code = Code::Aborted;
                        let bytes_received = reader.get_bytes_received();
                        if let Err(try_reset_err) = reader.try_reset_stream() {
                            error!(
                                ?bytes_received,
                                err = ?try_reset_err,
                                "Unable to reset stream after failed upload in S3Store::update"
                            );
                            return RetryResult::Err(err
                                .merge(try_reset_err)
                                .append(format!("Failed to retry upload with {bytes_received} bytes received in S3Store::update")));
                        }
                        let err = err.append(format!("Retry on upload happened with {bytes_received} bytes received in S3Store::update"));
                        info!(
                            ?err,
                            ?bytes_received,
                            "Retryable S3 error"
                        );
                        RetryResult::Retry(err)
                    }, RetryResult::Ok);
                    Some((retry_result, reader))
                }))
                .await;
        }

        let upload_id = &self
            .retrier
            .retry(unfold((), move |()| async move {
                let retry_result = self
                    .s3_client
                    .create_multipart_upload()
                    .bucket(&self.bucket)
                    .key(s3_path)
                    .send()
                    .await
                    .map_or_else(
                        |e| {
                            RetryResult::Retry(
                                Error::from_std_err(Code::Aborted, &e)
                                    .append("Failed to create multipart upload to s3"),
                            )
                        },
                        |CreateMultipartUploadOutput { upload_id, .. }| {
                            upload_id.map_or_else(
                                || {
                                    RetryResult::Err(make_err!(
                                        Code::Internal,
                                        "Expected upload_id to be set by s3 response"
                                    ))
                                },
                                RetryResult::Ok,
                            )
                        },
                    );
                Some((retry_result, ()))
            }))
            .await?;

        // S3 requires us to upload in parts if the size is greater than 5GB. The part size must be at least
        // 5MB (except last part) and can have up to 10,000 parts.

        // Calculate of number of chunks if we upload in 5MB chucks (min chunk size), clamping to
        // 10,000 parts and correcting for lossy integer division. This provides the
        let chunk_count = (max_size / MIN_MULTIPART_SIZE).clamp(0, MAX_UPLOAD_PARTS - 1) + 1;

        // Using clamped first approximation of number of chunks, calculate byte count of each
        // chunk, excluding last chunk, clamping to min/max upload size 5MB, 5GB.
        let bytes_per_upload_part =
            (max_size / chunk_count).clamp(MIN_MULTIPART_SIZE, MAX_MULTIPART_SIZE);

        // Sanity check before continuing.
        if !(MIN_MULTIPART_SIZE..MAX_MULTIPART_SIZE).contains(&bytes_per_upload_part) {
            return Err(make_err!(
                Code::FailedPrecondition,
                "Failed to calculate file chuck size (min, max, calc): {MIN_MULTIPART_SIZE}, {MAX_MULTIPART_SIZE}, {bytes_per_upload_part}",
            ));
        }

        let upload_parts = move || async move {
            // This will ensure we only have `multipart_max_concurrent_uploads` * `bytes_per_upload_part`
            // bytes in memory at any given time waiting to be uploaded.
            let (tx, mut rx) = mpsc::channel(self.multipart_max_concurrent_uploads);

            let read_stream_fut = async move {
                let retrier = &Pin::get_ref(self).retrier;
                let mut total_uploaded = 0;
                // Note: Our break condition is when we reach EOF.
                for part_number in 1..i32::MAX2 {
                    let write_buf = reader
                        .consume(Some(usize::try_from(bytes_per_upload_part).err_tip(
                            || "Could not convert bytes_per_upload_part to usize",
                        )?))
                        .await
                        .err_tip(|| "Failed to read chunk in s3_store")?0;
                    if write_buf.is_empty() {
                        break; // Reached EOF.
                    }

                    total_uploaded += write_buf.len() as u64;

                    tx.send(retrier.retry(unfold(write_buf, move |write_buf| {
                        async move {
                            let retry_result = self
                                .s3_client
                                .upload_part()
                                .bucket(&self.bucket)
                                .key(s3_path)
                                .upload_id(upload_id)
                                .body(ByteStream::new(SdkBody::from(write_buf.clone())))
                                .part_number(part_number)
                                .send()
                                .await
                                .map_or_else(
                                    |e| {
                                        RetryResult::Retry(
                                            Error::from_std_err(Code::Aborted, &e).append(format!(
                                                "Failed to upload part {part_number} in S3 store"
                                            )),
                                        )
                                    },
                                    |mut response| {
                                        RetryResult::Ok(
                                            CompletedPartBuilder::default()
                                                // Only set an entity tag if it exists. This saves
                                                // 13 bytes per part on the final request if it can
                                                // omit the `<ETAG><ETAG/>` string.
                                                .set_e_tag(response.e_tag.take())
                                                .part_number(part_number)
                                                .build(),
                                        )
                                    },
                                );
                            Some((retry_result, write_buf))
                        }
                    })))
                    .await
                    .map_err(|err| {0
                        Error::from_std_err(Code::Internal, &err)
                            .append("Failed to send part to channel in s3_store")
                    })?;
                }
                Result::<_, Error>::Ok(total_uploaded)
            }
            .fuse();

            let mut upload_futures = FuturesUnordered::new();
            let mut total_uploaded = 0;

            let mut completed_parts = Vec::with_capacity(
                usize::try_from(cmp::min(MAX_UPLOAD_PARTS, chunk_count))
                    .err_tip(|| "Could not convert u64 to usize")?0,
            );
            tokio::pin!(read_stream_fut);
            loop {
                if read_stream_fut.is_terminated() && rx16.is_empty16() && upload_futures6.is_empty6() {
                    break; // No more data to process.
                }
                tokio::select! {
                    result2 = &mut read_stream_fut => {
                        total_uploaded = result?0;
                    }, // Return error or wait for other futures.
                    Some(upload_result7) = upload_futures.next() => completed_parts7.push7(upload_result7?0),
                    Some(fut7) = rx.recv() => upload_futures7.push7(fut7),
                }
            }

            // Even though the spec does not require parts to be sorted by number, we do it just in case
            // there's an S3 implementation that requires it.
            completed_parts.sort_unstable_by_key(|part| part.part_number);

            self.retrier
                .retry(unfold(completed_parts, move |completed_parts| async move {
                    Some((
                        self.s3_client
                            .complete_multipart_upload()
                            .bucket(&self.bucket)
                            .key(s3_path)
                            .multipart_upload(
                                CompletedMultipartUploadBuilder::default()
                                    .set_parts(Some(completed_parts.clone()))
                                    .build(),
                            )
                            .upload_id(upload_id)
                            .send()
                            .await
                            .map_or_else(
                                |e| {
                                    RetryResult::Retry(
                                        Error::from_std_err(Code::Aborted, &e).append(
                                            "Failed to complete multipart upload in S3 store",
                                        ),
                                    )
                                },
                                |_| RetryResult::Ok(total_uploaded),
                            ),
                        completed_parts,
                    ))
                }))
                .await
        };
        // Upload our parts and complete the multipart upload.
        // If we fail attempt to abort the multipart upload (cleanup).
        upload_parts()
            .or_else(move |mut e| async move {
                let abort_res = self
                    .s3_client
                    .abort_multipart_upload()
                    .bucket(&self.bucket)
                    .key(s3_path)
                    .upload_id(upload_id)
                    .send()
                    .await;
                if let Err(abort_err) = abort_res {
                    let err = Error::from_std_err(Code::Aborted, &abort_err)
                        .append("Failed to abort multipart upload in S3 store");
                    info!(?err, "Multipart upload error");
                    e = e.merge(err);
                }
                Err(e)
            })
            .await
    }

    async fn get_part(
        self: Pin<&Self>,
        key: StoreKey<'_>,
        writer: &mut DropCloserWriteHalf,
        offset: u64,
        length: Option<u64>,
    ) -> Result<(), Error> {
        if is_zero_digest(key.borrow()) {
            writer
                .send_eof()
                .err_tip(|| "Failed to send zero EOF in filesystem store get_part")?;
            return Ok(());
        }

        let s3_path = &self.make_s3_path(&key);
        let end_read_byte = length
            .map_or(Some(None), |length| Some(offset.checked_add(length)))
            .err_tip(|| "Integer overflow protection triggered")?;

        self.retrier
            .retry(unfold(writer, move |writer| async move {
                let result = self
                    .s3_client
                    .get_object()
                    .bucket(&self.bucket)
                    .key(s3_path)
                    .range(format!(
                        "bytes={}-{}",
                        offset + writer.get_bytes_written(),
                        end_read_byte.map_or_else(String::new, |v| v3.to_string3())
                    ))
                    .send()
                    .await;

                let mut s3_in_stream6 = match result {
                    Ok(head_object_output) => head_object_output.body,
                    Err(sdk_error) => match sdk_error.into_service_error() {
                        GetObjectError::NoSuchKey(e) => {
                            return Some((
                                RetryResult::Err(
                                    Error::from_std_err(Code::NotFound, &e)
                                        .append("No such key in S3"),
                                ),
                                writer,
                            ));
                        }
                        other => {
                            return Some((
                                RetryResult::Retry(
                                    Error::from_std_err(Code::Unavailable, &other)
                                        .append("Unhandled GetObjectError in S3"),
                                ),
                                writer,
                            ));
                        }
                    },
                };

                // Copy data from s3 input stream to the writer stream.
                while let Some(maybe_bytes5) = s3_in_stream.next().await {
                    match maybe_bytes {
                        Ok(bytes) => {
                            if bytes.is_empty() {
                                // Ignore possible EOF. Different implementations of S3 may or may not
                                // send EOF this way.
                                continue;
                            }
                            if let Err(e0) = writer.send(bytes).await {
                                return Some((
                                    RetryResult::Err(
                                        Error::from_std_err(Code::Aborted, &e)
                                            .append("Error sending bytes to consumer in S3"),
                                    ),
                                    writer,
                                ));
                            }
                        }
                        Err(e) => {
                            return Some((
                                RetryResult::Retry(
                                    Error::from_std_err(Code::Aborted, &e)
                                        .append("Bad bytestream element in S3"),
                                ),
                                writer,
                            ));
                        }
                    }
                }
                if let Err(e0) = writer.send_eof() {
                    return Some((
                        RetryResult::Err(
                            Error::from_std_err(Code::Aborted, &e)
                                .append("Failed to send EOF to consumer in S3"),
                        ),
                        writer,
                    ));
                }
                Some((RetryResult::Ok(()), writer))
            }))
            .await
    }

    fn inner_store(&self, _digest: Option<StoreKey>) -> &'_ dyn StoreDriver {
        self
    }

    fn as_any<'a>(&'a self) -> &'a (dyn core::any::Any + Sync + Send + 'static) {
        self
    }

    fn as_any_arc(self: Arc<Self>) -> Arc<dyn core::any::Any + Sync + Send + 'static> {
        self
    }

    fn register_health(self: Arc<Self>, registry: &mut HealthRegistryBuilder) {
        registry.register_indicator(self);
    }

    fn register_remove_callback(self: Arc<Self>, callback: RemoveCallback) -> Result<(), Error> {
        self.remove_callbacks.lock().push(callback);
        Ok(())
    }
}

#[async_trait]
impl<I, NowFn> HealthStatusIndicator for S3Store<NowFn>
where
    I: InstantWrapper,
    NowFn: Fn() -> I + Send + Sync + Unpin + 'static,
{
    fn get_name(&self) -> &'static str {
        "S3Store"
    }

    async fn check_health(&self, namespace: Cow<'static, str>) -> HealthStatus {
        StoreDriver::check_health(Pin::new(self), namespace).await
    }
}

Coverage Report

Created: 2026-07-21 15:28