The “0-results” error I reported earlier has vanished on its own. Today’s test runs (2025-07-01 09:15 JST onward) are once again delivering the full 20 profiles with normal run times and pricing.

Although the Actor is functioning properly now, I’d like to prevent a repeat surprise. If possible, could you please advise on:

1. Automated health checks

   • Recommended indicators to monitor (e.g. dataset item count, run duration, exit code).
   • Options for alerting—webhooks, Slack, or Apify’s built-in task monitoring.

2. Backup / fallback strategy

   • Your preferred contingency if the Actor starts returning empty datasets again.

Sharing a checklist or example setup would be greatly appreciated. Thanks for your continued support!

Hi, thank you for the details! I apologize for the inconvenience. We are having an issue with LinkedIn blocking our requests due to the active usage from this actor and exceeding their rate limits. There are complications to scale it right now, but we're solving it with our supplier. Yesterday I added a log of "Too many requests" and disabled any charges when requests are being blocked. For example run https://console.apify.com/actors/M2FMdjRVeF1HPGFcc/runs/bQZI4gcVEllmWhda0#log

1. For now, an indicator is for example

const client = new ApifyClient({
  token: process.env.APIFY_API_KEY,
});
const run = await client.actor('harvestapi/linkedin-profile-search').call({...});

const { items } = await client.dataset(run.defaultDatasetId).listItems();

const logs = await client.log(run.id).get();
if (logs?.includes('Too many requests')) {
  ....
}

Let me look into other options.

2. Retry in several hours, when it's being blocked. And integration with an actor from another developer as a fallback.

Thank you for the quick follow-up and for adding the “Too many requests” safeguard (and the automatic no-charge logic)—that’s helpful! I’ll keep an eye on the log flag and set up alerts on my side as you suggested.

One idea that came up internally: would it be feasible to let users supply a list of personal li_at cookies and have the Actor automatically switch to the cookies whenever the current one hits LinkedIn’s rate limits? That kind of rotation might soften the blockage peaks while you work with your supplier on a broader fix. (In the other issue, you mentioned about the special account, so it may not be doable.)

Thanks again for the transparency and the rapid mitigation—looking forward to your thoughts!

It's doable via cookies, I totally agree it would be a good improvement or a fallback option. But this search won't work with normal accounts. We don't explicitly state here what kind of account, but I can give a hint. For example, this works the same way, (as I understand from the description, though I never tried it and I cannot vouch for it): <redacted>

It's an option to integrate with them (or any similar actor of your choice) as well, but be careful with Deep people profile extraction, as it may lead to your profile restriction when you scrape 500+ profiles per day. So I recommend using your cookies to scrape the search (as it's safer) and use our profile scraper https://apify.com/harvestapi/linkedin-profile-scraper to do the Deep profile extraction.

I’m sorry for the slow reply, and thank you for the detailed guidance!

I’ll explore the cookie-based approach you outlined. Since the underlying “zero-results” issue is no longer occurring for now, I’ll go ahead and close this issue for now.

Thanks again for your help and for the great Actor!

We are currently adding more accounts. It should work better. Thank you for understanding!