chore: bump version to 0.14.2 (#6032)

* chore: only retry when retry-able in flow (#5987)

* chore: only retry when retry-able

* chore: revert dbg change

* refactor: per review

* fix: check for available frontend first

* docs: more explain&longer timeout&feat: more retry at every level&try send select 1

* fix: use `sql` method for "SELECT 1"

* fix: also put recover flows in spawned task and a dead loop

* test: update transient error in flow rebuild test

* chore: sleep after sqlness sleep

* chore: add a warning

* chore: wait even more time after reboot

* fix: sanitize_connection_string (#6012)

* fix: disable recursion limit in prost (#6010)

Signed-off-by: Ruihang Xia <waynestxia@gmail.com>

* ci: fix the bugs of release-dev-builder-images and add update-dev-builder-image-tag (#6009)

* fix: the dev-builder release job is not triggered by merged event

* ci: add update-dev-builder-image-tag

* fix: always create mito engine (#6018)

* fix: force streaming mode for instant source table (#6031)

* fix: force streaming mode for instant source table

* tests: sqlness test&refactor: get table

* refactor: per review

* chore: bump version to 0.14.2

---------

Signed-off-by: Ruihang Xia <waynestxia@gmail.com>
Co-authored-by: jeremyhi <jiachun_feng@proton.me>
Co-authored-by: Ruihang Xia <waynestxia@gmail.com>
Co-authored-by: zyy17 <zyylsxm@gmail.com>
Co-authored-by: Lei, HUANG <6406592+v0y4g3r@users.noreply.github.com>
This commit is contained in:
discord9
2025-05-02 00:20:01 +08:00
committed by GitHub
parent e2df38d0d1
commit 5c9cbb5f4c
27 changed files with 935 additions and 223 deletions

View File

@@ -36,8 +36,8 @@ use common_grpc::flight::{FlightDecoder, FlightMessage};
use common_query::Output;
use common_recordbatch::error::ExternalSnafu;
use common_recordbatch::RecordBatchStreamWrapper;
use common_telemetry::error;
use common_telemetry::tracing_context::W3cTrace;
use common_telemetry::{error, warn};
use futures::future;
use futures_util::{Stream, StreamExt, TryStreamExt};
use prost::Message;
@@ -192,6 +192,36 @@ impl Database {
from_grpc_response(response)
}
/// Retry if connection fails, max_retries is the max number of retries, so the total wait time
/// is `max_retries * GRPC_CONN_TIMEOUT`
pub async fn handle_with_retry(&self, request: Request, max_retries: u32) -> Result<u32> {
let mut client = make_database_client(&self.client)?.inner;
let mut retries = 0;
let request = self.to_rpc_request(request);
loop {
let raw_response = client.handle(request.clone()).await;
match (raw_response, retries < max_retries) {
(Ok(resp), _) => return from_grpc_response(resp.into_inner()),
(Err(err), true) => {
// determine if the error is retryable
if is_grpc_retryable(&err) {
// retry
retries += 1;
warn!("Retrying {} times with error = {:?}", retries, err);
continue;
}
}
(Err(err), false) => {
error!(
"Failed to send request to grpc handle after {} retries, error = {:?}",
retries, err
);
return Err(err.into());
}
}
}
}
#[inline]
fn to_rpc_request(&self, request: Request) -> GreptimeRequest {
GreptimeRequest {
@@ -368,6 +398,11 @@ impl Database {
}
}
/// by grpc standard, only `Unavailable` is retryable, see: https://github.com/grpc/grpc/blob/master/doc/statuscodes.md#status-codes-and-their-use-in-grpc
pub fn is_grpc_retryable(err: &tonic::Status) -> bool {
matches!(err.code(), tonic::Code::Unavailable)
}
#[derive(Default, Debug, Clone)]
struct FlightContext {
auth_header: Option<AuthHeader>,