fix: Increase default navigation timeout to 30s, make it configurable…

… and add retries to crawling jobs
hoarder-app · Apr 11, 2024 · 238c296 · 238c296
1 parent be622e5
commit 238c296
Show file tree

Hide file tree

Showing 5 changed files with 17 additions and 6 deletions.
diff --git a/apps/workers/crawlerWorker.ts b/apps/workers/crawlerWorker.ts
@@ -170,7 +170,7 @@ async function crawlPage(jobId: string, url: string) {
     const page = await context.newPage();
 
     await page.goto(url, {
-      timeout: 10000, // 10 seconds
+      timeout: serverConfig.crawler.navigateTimeoutSec * 1000,
     });
     logger.info(
       `[Crawler][${jobId}] Successfully navigated to "${url}". Waiting for the page to load ...`,

diff --git a/apps/workers/utils.ts b/apps/workers/utils.ts
@@ -26,7 +26,6 @@ export async function readPDFText(buffer: Buffer): Promise<{
     const pdfParser = new PDFParser(null, 1);
     pdfParser.on("pdfParser_dataError", reject);
     pdfParser.on("pdfParser_dataReady", (pdfData) => {
-      // eslint-disable-next-line
       resolve({
         // The type isn't set correctly, reference : https://github.com/modesty/pdf2json/issues/327
         // eslint-disable-next-line

diff --git a/docs/docs/03-configuration.md b/docs/docs/03-configuration.md
@@ -37,6 +37,7 @@ Either `OPENAI_API_KEY` or `OLLAMA_BASE_URL` need to be set for automatic taggin
 
 ## Crawler Configs
 
-| Name                    | Required | Default | Description                                                                                                                                                            |
-| ----------------------- | -------- | ------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| CRAWLER_JOB_TIMEOUT_SEC | No       | 60      | How long to wait for the crawler job to finish before timing out. If you have a slow internet connection or a low powered device, you might want to bump this up a bit |
+| Name                         | Required | Default | Description                                                                                                                                                            |
+| ---------------------------- | -------- | ------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| CRAWLER_JOB_TIMEOUT_SEC      | No       | 60      | How long to wait for the crawler job to finish before timing out. If you have a slow internet connection or a low powered device, you might want to bump this up a bit |
+| CRAWLER_NAVIGATE_TIMEOUT_SEC | No       | 30      | How long to spend navigating to the page (along with its redirects). Increase this if you have a slow internet connection                                              |
diff --git a/packages/shared/config.ts b/packages/shared/config.ts
@@ -21,6 +21,7 @@ const allEnv = z.object({
   CRAWLER_HEADLESS_BROWSER: stringBool("true"),
   BROWSER_WEB_URL: z.string().url().optional(),
   CRAWLER_JOB_TIMEOUT_SEC: z.number().default(60),
+  CRAWLER_NAVIGATE_TIMEOUT_SEC: z.number().default(30),
   MEILI_ADDR: z.string().optional(),
   MEILI_MASTER_KEY: z.string().default(""),
   LOG_LEVEL: z.string().default("debug"),
@@ -58,6 +59,7 @@ const serverConfigSchema = allEnv.transform((val) => {
       headlessBrowser: val.CRAWLER_HEADLESS_BROWSER,
       browserWebUrl: val.BROWSER_WEB_URL,
       jobTimeoutSec: val.CRAWLER_JOB_TIMEOUT_SEC,
+      navigateTimeoutSec: val.CRAWLER_NAVIGATE_TIMEOUT_SEC,
     },
     meilisearch: val.MEILI_ADDR
       ? {

diff --git a/packages/shared/queues.ts b/packages/shared/queues.ts
@@ -17,7 +17,16 @@ export type ZCrawlLinkRequest = z.infer<typeof zCrawlLinkRequestSchema>;
 
 export const LinkCrawlerQueue = new Queue<ZCrawlLinkRequest, void>(
   "link_crawler_queue",
-  { connection: queueConnectionDetails },
+  {
+    connection: queueConnectionDetails,
+    defaultJobOptions: {
+      attempts: 5,
+      backoff: {
+        type: "exponential",
+        delay: 1000,
+      },
+    },
+  },
 );
 
 // OpenAI Worker