hubcio commented on code in PR #3523:
URL: https://github.com/apache/iggy/pull/3523#discussion_r3507330598
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
Review Comment:
`QuickwitSinkConfig` is missing `#[serde(deny_unknown_fields)]`. with every
knob optional, a typo'd key like `max_retires` is silently ignored and the
connector just runs on defaults with no error. influxdb_sink (the sink this
mirrors) sets it.
two smaller things on this struct: the `Serialize` derive is unused (the
config is only ever deserialized; the runtime serializes an untyped value, not
this type) so it can be dropped, and the retry/timeout knobs are undocumented
while http_sink documents each field with its default.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
Review Comment:
`max_retry_delay` reads backwards from the rest. influxdb, where this
startup-probe path is copied from, calls the same knob `retry_max_delay` and
pairs it with `open_retry_max_delay`. worth renaming the field + the
`DEFAULT_MAX_RETRY_DELAY` const to match, since config keys turn into a compat
contract once this ships.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
let reason = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when creating index: {}.
Status code: {status}, reason: {reason}",
+ "Permanent client error creating Quickwit index: {} for
connector ID: {}. status: {status}, reason: {reason}",
+ self.index_id, self.id
+ );
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
+ )))
+ } else {
+ let reason = response.text().await.unwrap_or_default();
+ error!(
+ "Server error creating Quickwit index: {} for connector ID:
{}. status: {status}, reason: {reason}",
+ self.index_id, self.id
);
- return Err(Error::InitError(format!(
- "Failed to create index: {}. {reason}",
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
- )));
+ )))
}
-
- info!("Created index: {}", self.index_id);
- Ok(())
}
pub async fn ingest(&self, messages: Vec<simd_json::OwnedValue>) ->
Result<(), Error> {
+ let client = self.client()?;
+ // At-least-once: Quickwit ingest carries no dedup key, so a retry
after a
Review Comment:
this comment only covers the duplicate-write window. it's silent on the loss
window: on a permanent 4xx, or once retries are exhausted, the offset was
already committed at poll, so the batch is dropped and never redelivered. worth
stating both so the real guarantee (at-most-once on final failure) is clear.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
let reason = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when creating index: {}.
Status code: {status}, reason: {reason}",
+ "Permanent client error creating Quickwit index: {} for
connector ID: {}. status: {status}, reason: {reason}",
+ self.index_id, self.id
+ );
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
+ )))
+ } else {
+ let reason = response.text().await.unwrap_or_default();
+ error!(
+ "Server error creating Quickwit index: {} for connector ID:
{}. status: {status}, reason: {reason}",
+ self.index_id, self.id
);
- return Err(Error::InitError(format!(
- "Failed to create index: {}. {reason}",
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
- )));
+ )))
}
-
- info!("Created index: {}", self.index_id);
- Ok(())
}
pub async fn ingest(&self, messages: Vec<simd_json::OwnedValue>) ->
Result<(), Error> {
+ let client = self.client()?;
+ // At-least-once: Quickwit ingest carries no dedup key, so a retry
after a
+ // 5xx/timeout that actually committed (commit=auto) double-writes
those rows.
let url = format!(
Review Comment:
these urls (here, plus has_index and create_index) are rebuilt with
`format!` on every batch, and none trim a trailing slash - a url ending in `/`
produces `//api/v1/...`. build them once in `open()` and
`trim_end_matches('/')` the base, like influxdb's `base_url()`.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
let reason = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when creating index: {}.
Status code: {status}, reason: {reason}",
+ "Permanent client error creating Quickwit index: {} for
connector ID: {}. status: {status}, reason: {reason}",
+ self.index_id, self.id
+ );
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
+ )))
+ } else {
+ let reason = response.text().await.unwrap_or_default();
+ error!(
+ "Server error creating Quickwit index: {} for connector ID:
{}. status: {status}, reason: {reason}",
+ self.index_id, self.id
);
- return Err(Error::InitError(format!(
- "Failed to create index: {}. {reason}",
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
- )));
+ )))
}
-
- info!("Created index: {}", self.index_id);
- Ok(())
}
pub async fn ingest(&self, messages: Vec<simd_json::OwnedValue>) ->
Result<(), Error> {
+ let client = self.client()?;
+ // At-least-once: Quickwit ingest carries no dedup key, so a retry
after a
+ // 5xx/timeout that actually committed (commit=auto) double-writes
those rows.
let url = format!(
"{}/api/v1/{}/ingest?commit=auto",
self.config.url, self.index_id
);
- info!("Ingesting messages for index: {}...", self.index_id);
let messages_count = messages.len();
- let messages = messages
- .into_iter()
- .filter_map(|record| simd_json::to_string(&record).ok())
- .collect::<Vec<_>>()
- .join("\n");
-
- let response = self
- .client
+ let mut ndjson = String::new();
Review Comment:
ndjson starts from `String::new()` and reallocs as it grows over the batch
(up to batch_length records). pre-size it with `String::with_capacity(...)` -
the commit says "optimize allocations" but this is the main alloc site.
separately: records where `to_string` fails get skipped silently while
`messages_count` still counts them, so the success log can overcount.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
let reason = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when creating index: {}.
Status code: {status}, reason: {reason}",
+ "Permanent client error creating Quickwit index: {} for
connector ID: {}. status: {status}, reason: {reason}",
+ self.index_id, self.id
+ );
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
+ )))
+ } else {
+ let reason = response.text().await.unwrap_or_default();
+ error!(
+ "Server error creating Quickwit index: {} for connector ID:
{}. status: {status}, reason: {reason}",
+ self.index_id, self.id
);
- return Err(Error::InitError(format!(
- "Failed to create index: {}. {reason}",
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
- )));
+ )))
}
-
- info!("Created index: {}", self.index_id);
- Ok(())
}
pub async fn ingest(&self, messages: Vec<simd_json::OwnedValue>) ->
Result<(), Error> {
+ let client = self.client()?;
+ // At-least-once: Quickwit ingest carries no dedup key, so a retry
after a
+ // 5xx/timeout that actually committed (commit=auto) double-writes
those rows.
let url = format!(
"{}/api/v1/{}/ingest?commit=auto",
self.config.url, self.index_id
);
- info!("Ingesting messages for index: {}...", self.index_id);
let messages_count = messages.len();
- let messages = messages
- .into_iter()
- .filter_map(|record| simd_json::to_string(&record).ok())
- .collect::<Vec<_>>()
- .join("\n");
-
- let response = self
- .client
+ let mut ndjson = String::new();
+ for record in messages {
+ if let Ok(json_str) = simd_json::to_string(&record) {
+ if !ndjson.is_empty() {
+ ndjson.push('\n');
+ }
+ ndjson.push_str(&json_str);
+ }
+ }
+
+ let response = client
.post(&url)
- .body(messages)
+ .header("Content-Type", "application/x-ndjson")
+ .body(ndjson)
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to ingest messages for index:
{}. {error}",
- self.index_id
+ "Failed to ingest {messages_count} messages into Quickwit
index: {} for connector ID: {}. {e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ debug!(
+ "Ingested {messages_count} messages into Quickwit index: {}
for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() && status !=
StatusCode::TOO_MANY_REQUESTS {
let text = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when ingesting messages for
index: {}. Status code: {status}, reason: {text}",
- self.index_id
+ "Permanent error ingesting into Quickwit index: {} for
connector ID: {}. status: {status}, reason: {text}",
+ self.index_id, self.id
+ );
+ Err(Error::PermanentHttpError(format!(
Review Comment:
this permanent-vs-transient split has no runtime effect: the error is
collapsed to a return code at the ffi boundary and the runtime discards it, and
this sink owns no circuit breaker - so the description's "so the circuit
breaker is not tripped" doesn't apply here. the two distinct log lines are
still handy for triage, keep those; it's just the error variant that doesn't
gate anything.
##########
core/connectors/sinks/quickwit_sink/Cargo.toml:
##########
@@ -31,15 +31,15 @@ publish = false
[package.metadata.cargo-machete]
Review Comment:
this `ignored = ["dashmap"]` is dangling now - the pr drops the dashmap
dependency, so cargo-machete has nothing to ignore here anymore. delete the
stanza. the macro's `DashMap` comes from the sdk re-export, not a direct dep.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
let reason = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when creating index: {}.
Status code: {status}, reason: {reason}",
+ "Permanent client error creating Quickwit index: {} for
connector ID: {}. status: {status}, reason: {reason}",
+ self.index_id, self.id
+ );
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
+ )))
+ } else {
+ let reason = response.text().await.unwrap_or_default();
+ error!(
+ "Server error creating Quickwit index: {} for connector ID:
{}. status: {status}, reason: {reason}",
+ self.index_id, self.id
);
- return Err(Error::InitError(format!(
- "Failed to create index: {}. {reason}",
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
- )));
+ )))
}
-
- info!("Created index: {}", self.index_id);
- Ok(())
}
pub async fn ingest(&self, messages: Vec<simd_json::OwnedValue>) ->
Result<(), Error> {
+ let client = self.client()?;
+ // At-least-once: Quickwit ingest carries no dedup key, so a retry
after a
+ // 5xx/timeout that actually committed (commit=auto) double-writes
those rows.
let url = format!(
"{}/api/v1/{}/ingest?commit=auto",
self.config.url, self.index_id
);
- info!("Ingesting messages for index: {}...", self.index_id);
let messages_count = messages.len();
- let messages = messages
- .into_iter()
- .filter_map(|record| simd_json::to_string(&record).ok())
- .collect::<Vec<_>>()
- .join("\n");
-
- let response = self
- .client
+ let mut ndjson = String::new();
+ for record in messages {
+ if let Ok(json_str) = simd_json::to_string(&record) {
+ if !ndjson.is_empty() {
+ ndjson.push('\n');
+ }
+ ndjson.push_str(&json_str);
+ }
+ }
+
+ let response = client
.post(&url)
- .body(messages)
+ .header("Content-Type", "application/x-ndjson")
+ .body(ndjson)
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to ingest messages for index:
{}. {error}",
- self.index_id
+ "Failed to ingest {messages_count} messages into Quickwit
index: {} for connector ID: {}. {e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ debug!(
+ "Ingested {messages_count} messages into Quickwit index: {}
for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() && status !=
StatusCode::TOO_MANY_REQUESTS {
let text = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when ingesting messages for
index: {}. Status code: {status}, reason: {text}",
- self.index_id
+ "Permanent error ingesting into Quickwit index: {} for
connector ID: {}. status: {status}, reason: {text}",
+ self.index_id, self.id
+ );
+ Err(Error::PermanentHttpError(format!(
+ "status: {status}, reason: {text}"
+ )))
+ } else {
+ let text = response.text().await.unwrap_or_default();
+ error!(
+ "Transient error ingesting into Quickwit index: {} for
connector ID: {}. status: {status}, reason: {text}",
+ self.index_id, self.id
);
- return Err(Error::HttpRequestFailed(format!(
- "Status code: {status}, reason: {text}"
- )));
+ Err(Error::HttpRequestFailed(format!(
+ "status: {status}, reason: {text}"
+ )))
}
+ }
- info!(
- "Ingested {messages_count} messages for index: {}",
- self.index_id
- );
- Ok(())
+ fn extract_json_payloads(
+ &self,
+ messages: Vec<ConsumedMessage>,
+ schema: Schema,
+ ) -> Vec<OwnedValue> {
+ let mut json_payloads = Vec::with_capacity(messages.len());
+ for message in messages {
+ let val = match message.payload {
+ Payload::Json(value) => value,
+ Payload::Raw(bytes) => {
+ let mut bytes_copy = bytes.clone();
+ match simd_json::from_slice::<OwnedValue>(&mut bytes_copy)
{
+ Ok(value) => value,
+ Err(_) => {
+ if let Ok(text) = String::from_utf8(bytes.clone())
{
+ simd_json::json!({
+ "data": text,
+ "data_type": "raw"
+ })
+ } else {
+ simd_json::json!({
+ "data":
general_purpose::STANDARD.encode(&bytes),
+ "data_type": "raw"
+ })
+ }
+ }
+ }
+ }
+ Payload::Text(text) => simd_json::json!({
+ "text": text,
+ "data_type": "text"
+ }),
+ _ => {
+ warn!(
+ "Quickwit sink connector ID: {} unsupported payload
schema: {}",
+ self.id, schema
+ );
+ continue;
+ }
+ };
+ json_payloads.push(val);
+ }
+ json_payloads
}
}
#[async_trait]
impl Sink for QuickwitSink {
async fn open(&mut self) -> Result<(), Error> {
- info!(
- "Opened Quickwit sink connector with ID: {} for URL: {}",
- self.id, self.config.url
+ let parsed: IndexIdExtract =
serde_yaml_ng::from_str(&self.config.index)
+ .map_err(|e| Error::InvalidConfigValue(format!("index: invalid
YAML: {e}")))?;
+ self.index_id = parsed.index_id;
+
+ let retry_delay = parse_duration(self.config.retry_delay.as_deref(),
DEFAULT_RETRY_DELAY);
+ let max_retry_delay = parse_duration(
+ self.config.max_retry_delay.as_deref(),
+ DEFAULT_MAX_RETRY_DELAY,
+ );
+ let max_open_retries = self
+ .config
+ .max_open_retries
+ .unwrap_or(DEFAULT_MAX_OPEN_RETRIES);
+ let open_retry_max_delay = parse_duration(
+ self.config.open_retry_max_delay.as_deref(),
+ DEFAULT_OPEN_RETRY_MAX_DELAY,
);
+
+ let request_timeout = parse_duration(
+ self.config.request_timeout.as_deref(),
+ DEFAULT_REQUEST_TIMEOUT,
+ );
+ let raw_client = reqwest::Client::builder()
+ .timeout(request_timeout)
+ .build()
+ .map_err(|e| Error::InitError(format!("reqwest client: {e}")))?;
+ let health_url = Url::parse(&format!("{}/health/livez",
self.config.url))
Review Comment:
`/health/livez` is liveness - the node can be live but not yet ready to
serve. `/health/readyz` is the readiness gate you actually want before
create_index/ingest. minor, since the retry client papers over an early 5xx,
but readyz is the more correct probe.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
Review Comment:
every other http sink names this `timeout` (influxdb, http_sink, doris) -
`request_timeout` is the lone outlier. rename the field +
`DEFAULT_REQUEST_TIMEOUT` const before release, config keys are hard to change
later.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
Review Comment:
these two arms build an identical `Err(InitError(...))` and differ only in
the log wording, and the status is already in the message. collapse them into
one else. keep the success and 409 arms separate - the 409-absorb is
load-bearing for the create race.
##########
core/connectors/sinks/quickwit_sink/src/lib.rs:
##########
@@ -16,158 +16,325 @@
// under the License.
use async_trait::async_trait;
+use base64::{Engine as _, engine::general_purpose};
+use iggy_connector_sdk::retry::{
+ ConnectivityConfig, build_retry_client, check_connectivity_with_retry,
parse_duration,
+};
use iggy_connector_sdk::{
- ConsumedMessage, Error, MessagesMetadata, Payload, Sink, TopicMetadata,
sink_connector,
+ ConsumedMessage, Error, MessagesMetadata, Payload, Schema, Sink,
TopicMetadata, sink_connector,
};
+use reqwest::StatusCode;
+use reqwest::Url;
+use reqwest_middleware::ClientWithMiddleware;
use serde::{Deserialize, Serialize};
-use tracing::{error, info, warn};
+use simd_json::OwnedValue;
+use tracing::{debug, error, info, warn};
sink_connector!(QuickwitSink);
+const DEFAULT_MAX_RETRIES: u32 = 3;
+const DEFAULT_RETRY_DELAY: &str = "200ms";
+const DEFAULT_MAX_RETRY_DELAY: &str = "5s";
+const DEFAULT_MAX_OPEN_RETRIES: u32 = 5;
+const DEFAULT_OPEN_RETRY_MAX_DELAY: &str = "30s";
+const DEFAULT_REQUEST_TIMEOUT: &str = "30s";
+
#[derive(Debug)]
pub struct QuickwitSink {
id: u32,
config: QuickwitSinkConfig,
- client: reqwest::Client,
+ client: Option<ClientWithMiddleware>,
+ verbose: bool,
index_id: String,
}
#[derive(Debug, Serialize, Deserialize)]
pub struct QuickwitSinkConfig {
- url: String,
- index: String,
+ pub url: String,
+ /// Full Quickwit index config YAML, passed to `POST /api/v1/indexes` on
first open.
+ /// `index_id` is extracted from this YAML to build ingest URLs.
+ pub index: String,
+ pub verbose_logging: Option<bool>,
+ pub max_retries: Option<u32>,
+ pub retry_delay: Option<String>,
+ pub max_retry_delay: Option<String>,
+ pub max_open_retries: Option<u32>,
+ pub open_retry_max_delay: Option<String>,
+ pub request_timeout: Option<String>,
}
-#[derive(Debug, Serialize, Deserialize)]
-struct IndexConfig {
+#[derive(Debug, Deserialize)]
+struct IndexIdExtract {
index_id: String,
}
impl QuickwitSink {
pub fn new(id: u32, config: QuickwitSinkConfig) -> Self {
- let index_config =
-
serde_yaml_ng::from_str::<IndexConfig>(&config.index).expect("Invalid index
config.");
- QuickwitSink {
+ let verbose = config.verbose_logging.unwrap_or(false);
+ Self {
id,
config,
- index_id: index_config.index_id,
- client: reqwest::Client::new(),
+ client: None,
+ verbose,
+ index_id: String::new(),
}
}
+ fn client(&self) -> Result<&ClientWithMiddleware, Error> {
+ self.client
+ .as_ref()
+ .ok_or_else(|| Error::InitError("Quickwit sink client not
initialized".into()))
+ }
+
async fn has_index(&self) -> Result<bool, Error> {
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes/{}", self.config.url,
self.index_id);
- let response = self.client.get(&url).send().await.map_err(|error| {
- error!(
- "Failed to send HTTP request to check if index with ID: {}
exists. {error}",
- self.index_id
- );
- Error::HttpRequestFailed(error.to_string())
- })?;
+ let response = client
+ .get(&url)
+ .send()
+ .await
+ .map_err(|e| Error::HttpRequestFailed(e.to_string()))?;
let status = response.status();
if status.is_success() {
Ok(true)
- } else if status == reqwest::StatusCode::NOT_FOUND {
+ } else if status == StatusCode::NOT_FOUND {
Ok(false)
} else {
Err(Error::HttpRequestFailed(format!(
- "Unexpected status code: {status}",
+ "Unexpected status checking Quickwit index: {status}"
)))
}
}
async fn create_index(&self) -> Result<(), Error> {
- info!("Creating index: {}", self.index_id);
+ info!(
+ "Creating Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ let client = self.client()?;
let url = format!("{}/api/v1/indexes", self.config.url);
- let response = self
- .client
+ let response = client
.post(&url)
- .header("content-type", "application/yaml")
- .body(self.config.index.to_owned())
+ .header("Content-Type", "application/yaml")
+ .body(self.config.index.clone())
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to create index: {}. {error}",
- self.index_id
+ "Failed to create Quickwit index: {} for connector ID: {}.
{e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ info!(
+ "Created Quickwit index: {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status == StatusCode::CONFLICT {
+ // Another instance beat us to it; the index exists, which is what
we want.
+ info!(
+ "Quickwit index already exists (409): {} for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() {
let reason = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when creating index: {}.
Status code: {status}, reason: {reason}",
+ "Permanent client error creating Quickwit index: {} for
connector ID: {}. status: {status}, reason: {reason}",
+ self.index_id, self.id
+ );
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
+ )))
+ } else {
+ let reason = response.text().await.unwrap_or_default();
+ error!(
+ "Server error creating Quickwit index: {} for connector ID:
{}. status: {status}, reason: {reason}",
+ self.index_id, self.id
);
- return Err(Error::InitError(format!(
- "Failed to create index: {}. {reason}",
+ Err(Error::InitError(format!(
+ "Failed to create index '{0}': {status} {reason}",
self.index_id
- )));
+ )))
}
-
- info!("Created index: {}", self.index_id);
- Ok(())
}
pub async fn ingest(&self, messages: Vec<simd_json::OwnedValue>) ->
Result<(), Error> {
+ let client = self.client()?;
+ // At-least-once: Quickwit ingest carries no dedup key, so a retry
after a
+ // 5xx/timeout that actually committed (commit=auto) double-writes
those rows.
let url = format!(
"{}/api/v1/{}/ingest?commit=auto",
self.config.url, self.index_id
);
- info!("Ingesting messages for index: {}...", self.index_id);
let messages_count = messages.len();
- let messages = messages
- .into_iter()
- .filter_map(|record| simd_json::to_string(&record).ok())
- .collect::<Vec<_>>()
- .join("\n");
-
- let response = self
- .client
+ let mut ndjson = String::new();
+ for record in messages {
+ if let Ok(json_str) = simd_json::to_string(&record) {
+ if !ndjson.is_empty() {
+ ndjson.push('\n');
+ }
+ ndjson.push_str(&json_str);
+ }
+ }
+
+ let response = client
.post(&url)
- .body(messages)
+ .header("Content-Type", "application/x-ndjson")
+ .body(ndjson)
.send()
.await
- .map_err(|error| {
+ .map_err(|e| {
error!(
- "Failed to send HTTP request to ingest messages for index:
{}. {error}",
- self.index_id
+ "Failed to ingest {messages_count} messages into Quickwit
index: {} for connector ID: {}. {e}",
+ self.index_id, self.id
);
- Error::HttpRequestFailed(error.to_string())
+ Error::HttpRequestFailed(e.to_string())
})?;
- if !response.status().is_success() {
- let status = response.status();
+ let status = response.status();
+ if status.is_success() {
+ debug!(
+ "Ingested {messages_count} messages into Quickwit index: {}
for connector ID: {}",
+ self.index_id, self.id
+ );
+ Ok(())
+ } else if status.is_client_error() && status !=
StatusCode::TOO_MANY_REQUESTS {
let text = response.text().await.unwrap_or_default();
error!(
- "Received an invalid HTTP response when ingesting messages for
index: {}. Status code: {status}, reason: {text}",
- self.index_id
+ "Permanent error ingesting into Quickwit index: {} for
connector ID: {}. status: {status}, reason: {text}",
+ self.index_id, self.id
+ );
+ Err(Error::PermanentHttpError(format!(
+ "status: {status}, reason: {text}"
+ )))
+ } else {
+ let text = response.text().await.unwrap_or_default();
+ error!(
+ "Transient error ingesting into Quickwit index: {} for
connector ID: {}. status: {status}, reason: {text}",
+ self.index_id, self.id
);
- return Err(Error::HttpRequestFailed(format!(
- "Status code: {status}, reason: {text}"
- )));
+ Err(Error::HttpRequestFailed(format!(
+ "status: {status}, reason: {text}"
+ )))
}
+ }
- info!(
- "Ingested {messages_count} messages for index: {}",
- self.index_id
- );
- Ok(())
+ fn extract_json_payloads(
+ &self,
+ messages: Vec<ConsumedMessage>,
+ schema: Schema,
+ ) -> Vec<OwnedValue> {
+ let mut json_payloads = Vec::with_capacity(messages.len());
+ for message in messages {
+ let val = match message.payload {
+ Payload::Json(value) => value,
+ Payload::Raw(bytes) => {
+ let mut bytes_copy = bytes.clone();
+ match simd_json::from_slice::<OwnedValue>(&mut bytes_copy)
{
+ Ok(value) => value,
+ Err(_) => {
+ if let Ok(text) = String::from_utf8(bytes.clone())
{
Review Comment:
`String::from_utf8(bytes.clone())` clones a second time. `from_utf8(bytes)`
can consume the buffer, and on the error path `e.into_bytes()` hands it back
for the base64 branch. the earlier clone at the parse call is load-bearing
(simd parses destructively) so keep that one.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
