Skip to content

Fixed retrying on run replication inserts on connection errors #2216

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 1, 2025
Merged

Fixed retrying on run replication inserts on connection errors #2216

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 15 additions & 10 deletions apps/webapp/app/services/runsReplicationService.server.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -462,13 +462,13 @@ export class RunsReplicationService {

// Insert task runs and payloads with retry logic for connection errors
const [taskRunError, taskRunResult] = await this.#insertWithRetry(
() => this.#insertTaskRunInserts(taskRunInserts),
(attempt) => this.#insertTaskRunInserts(taskRunInserts, attempt),
"task run inserts",
flushId
);

const [payloadError, payloadResult] = await this.#insertWithRetry(
() => this.#insertPayloadInserts(payloadInserts),
(attempt) => this.#insertPayloadInserts(payloadInserts, attempt),
"payload inserts",
flushId
);
Expand Down Expand Up @@ -502,24 +502,25 @@ export class RunsReplicationService {

// New method to handle inserts with retry logic for connection errors
async #insertWithRetry<T>(
insertFn: () => Promise<T>,
insertFn: (attempt: number) => Promise<T>,
operationName: string,
flushId: string
): Promise<[Error | null, T | null]> {
let lastError: Error | null = null;

for (let attempt = 1; attempt <= this._insertMaxRetries; attempt++) {
try {
const result = await insertFn();
const result = await insertFn(attempt);
return [null, result];
} catch (error) {
lastError = error instanceof Error ? error : new Error(String(error));

// Check if this is a retryable connection error
if (this.#isRetryableConnectionError(lastError) && attempt < this._insertMaxRetries) {
if (this.#isRetryableConnectionError(lastError)) {
const delay = this.#calculateConnectionRetryDelay(attempt);

this.logger.warn(`Retrying ${operationName} due to connection error`, {
this.logger.warn(`Retrying RunReplication insert due to connection error`, {
operationName,
flushId,
attempt,
maxRetries: this._insertMaxRetries,
Expand Down Expand Up @@ -567,7 +568,7 @@ export class RunsReplicationService {
return delay + jitter;
}

async #insertTaskRunInserts(taskRunInserts: TaskRunV2[]) {
async #insertTaskRunInserts(taskRunInserts: TaskRunV2[], attempt: number) {
return await startSpan(this._tracer, "insertTaskRunsInserts", async (span) => {
const [insertError, insertResult] = await this.options.clickhouse.taskRuns.insert(
taskRunInserts,
Expand All @@ -581,18 +582,20 @@ export class RunsReplicationService {
);

if (insertError) {
this.logger.error("Error inserting task run inserts", {
this.logger.error("Error inserting task run inserts attempt", {
error: insertError,
attempt,
});

recordSpanError(span, insertError);
throw insertError;
}

return insertResult;
});
}

async #insertPayloadInserts(payloadInserts: RawTaskRunPayloadV1[]) {
async #insertPayloadInserts(payloadInserts: RawTaskRunPayloadV1[], attempt: number) {
return await startSpan(this._tracer, "insertPayloadInserts", async (span) => {
const [insertError, insertResult] = await this.options.clickhouse.taskRuns.insertPayloads(
payloadInserts,
Expand All @@ -606,11 +609,13 @@ export class RunsReplicationService {
);

if (insertError) {
this.logger.error("Error inserting payload inserts", {
this.logger.error("Error inserting payload inserts attempt", {
error: insertError,
attempt,
});

recordSpanError(span, insertError);
throw insertError;
}

return insertResult;
Expand Down