Skip to content

🐸 Rybbit - open-source and privacy-friendly alternative to Google Analytics that is 10x more intuitive.

Simplify error handling

Edward Cui6be21b577523
diff --git a/server/src/services/import/workers/csvParseWorker.ts b/server/src/services/import/workers/csvParseWorker.ts
index cd7d7ddaa..d0e8fcfad 100644
--- a/server/src/services/import/workers/csvParseWorker.ts
+++ b/server/src/services/import/workers/csvParseWorker.ts
@@ -22,6 +22,14 @@ const getImportDataHeaders = (platform: string) => {
}
};
+const safeUpdateStatusToFailed = async (importId: string, errorMessage: string) => {
+ try {
+ await updateImportStatus(importId, "failed", errorMessage);
+ } catch (updateError) {
+ logger.error({ importId, error: updateError }, "Could not update status to failed");
+ }
+};
+
const createR2FileStream = async (storageLocation: string, platform: string) => {
logger.info({ storageLocation }, "Reading from R2");
const fileStream = await r2Storage.getImportFileStream(storageLocation);
@@ -89,7 +97,7 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
const quotaTracker = await ImportQuotaTracker.create(organization);
const chunkSize = 5000;
- const INACTIVITY_TIMEOUT_MS = 90 * 1000; // 90 seconds of inactivity
+ const INACTIVITY_TIMEOUT_MS = 90 * 1000;
let chunk: UmamiEvent[] = [];
let totalSkippedQuota = 0;
@@ -99,16 +107,13 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
? await createR2FileStream(storageLocation, platform)
: await createLocalFileStream(storageLocation, platform);
- // Add explicit error handler before starting to consume the stream
stream.on("error", error => {
logger.error({ importId, error }, "Stream error");
- // Error will be caught by the outer try/catch
});
await updateImportStatus(importId, "processing");
logger.info({ importId, platform, organization }, "Started processing CSV import");
- // Helper to reset inactivity timeout - resets whenever progress is made
const resetTimeout = () => {
if (processingTimeout) {
clearTimeout(processingTimeout);
@@ -142,7 +147,6 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
continue;
}
- // Event passed all filters - add to chunk
chunk.push(data);
totalProcessed++;
@@ -155,12 +159,10 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
allChunksSent: false,
});
chunk = [];
- // Reset timeout after successful chunk send
resetTimeout();
}
}
- // Send final chunk if any data remains
if (chunk.length > 0) {
await jobQueue.send<DataInsertJob>(DATA_INSERT_QUEUE, {
site,
@@ -169,11 +171,9 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
chunk,
allChunksSent: false,
});
- // Reset timeout after final chunk send
resetTimeout();
}
- // Check if some events couldn't be imported due to quotas
if (totalSkippedQuota > 0) {
const quotaSummary = quotaTracker.getSummary();
const errorMessage =
@@ -188,7 +188,6 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
return;
}
- // Send finalization signal with total chunk count
logger.info({ importId, totalProcessed, totalSkippedQuota }, "CSV parsing completed successfully");
await jobQueue.send<DataInsertJob>(DATA_INSERT_QUEUE, {
site,
@@ -199,18 +198,13 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
});
} catch (error) {
logger.error({ importId, error }, "Error in CSV parse worker");
-
- await updateImportStatus(importId, "failed", "An unexpected error occurred during import processing");
-
- // Don't re-throw - worker should continue processing other jobs
+ await safeUpdateStatusToFailed(importId, "An unexpected error occurred during import processing");
logger.error({ importId }, "Import failed, worker continuing");
} finally {
- // Clean up timeout
if (processingTimeout) {
clearTimeout(processingTimeout);
}
- // Ensure stream is destroyed to prevent memory leaks
if (stream) {
try {
stream.destroy();
@@ -219,12 +213,16 @@ export async function createCsvParseWorker(jobQueue: IJobQueue) {
}
}
- // Clean up file - don't throw on failure to prevent worker crashes
- const deleteResult = await deleteImportFile(storageLocation, isR2Storage);
- if (!deleteResult.success) {
- logger.warn({ importId, error: deleteResult.error }, "File cleanup failed, will remain in storage");
- // File will be orphaned but import status is already recorded
- // importCleanupService.ts handles orphans
+ // Clean up file - log errors but don't throw to prevent worker crashes
+ try {
+ const deleteResult = await deleteImportFile(storageLocation, isR2Storage);
+ if (!deleteResult.success) {
+ logger.warn({ importId, error: deleteResult.error }, "File cleanup failed, will remain in storage");
+ // File will be orphaned but import status is already recorded
+ // importCleanupService.ts handles orphans
+ }
+ } catch (deleteError) {
+ logger.error({ importId, error: deleteError }, "Critical error during file cleanup");
}
}
});
diff --git a/server/src/services/import/workers/dataInsertWorker.ts b/server/src/services/import/workers/dataInsertWorker.ts
index ef311f1a6..cdab8a131 100644
--- a/server/src/services/import/workers/dataInsertWorker.ts
+++ b/server/src/services/import/workers/dataInsertWorker.ts
@@ -16,33 +16,28 @@ const getImportDataMapping = (platform: string) => {
}
};
+const safeUpdateStatusToFailed = async (importId: string, errorMessage: string) => {
+ try {
+ await updateImportStatus(importId, "failed", errorMessage);
+ } catch (updateError) {
+ logger.error({ importId, error: updateError }, "Could not update status to failed");
+ }
+};
+
export async function createDataInsertWorker(jobQueue: IJobQueue) {
await jobQueue.work<DataInsertJob>(DATA_INSERT_QUEUE, async job => {
const { site, importId, platform, chunk, allChunksSent } = job;
- if (allChunksSent) {
- try {
+ try {
+ if (allChunksSent) {
await updateImportStatus(importId, "completed");
logger.info({ importId }, "Import completed successfully");
return;
- } catch (error) {
- logger.error({ importId, error }, "Failed to mark as completed");
- // Try to update to failed status, but don't crash worker
- try {
- await updateImportStatus(importId, "failed", "Failed to complete import");
- } catch (updateError) {
- logger.error({ importId, error: updateError }, "Could not update status to failed");
- }
- // Don't re-throw - worker should continue
- return;
}
- }
- try {
const dataMapper = getImportDataMapping(platform);
const transformedRecords = dataMapper.transform(chunk, site, importId);
- // Insert to ClickHouse (critical - must succeed)
await clickhouse.insert({
table: "events",
values: transformedRecords,
@@ -62,16 +57,12 @@ export async function createDataInsertWorker(jobQueue: IJobQueue) {
// Don't throw - data is safely in ClickHouse, progress can be off slightly
}
} catch (error) {
- logger.error({ importId, error }, "ClickHouse insert failed");
+ logger.error({ importId, error }, "Worker processing failed");
- try {
- await updateImportStatus(importId, "failed", "Data insertion failed due to unknown error");
- } catch (updateError) {
- logger.error({ importId, error: updateError }, "Could not update status to failed");
- }
+ const errorMessage = allChunksSent ? "Failed to complete import" : "Data insertion failed due to unknown error";
+ await safeUpdateStatusToFailed(importId, errorMessage);
- // Don't re-throw - worker should continue processing other jobs
- logger.error({ importId }, "Import chunk failed, worker continuing");
+ logger.error({ importId }, "Import operation failed, worker continuing");
}
});
}