Published by NewsPR Today | September 2025
Introduction
Google PageSpeed Insights is a popular tool for assessing the performance of websites. It gives marketers and developers data-driven insights into user experience, SEO, accessibility, and site speed, all powered by the Lighthouse engine. However, users sometimes run audits and get unexpected errors, especially when it comes to memory limits and throttling.
One such issue, flagged by messages like extensible_stubs::UNABLE_TO_RETRY and THROTTLED_TASK_LIMIT, has raised concerns within the web development community. These errors can interrupt performance testing and delay optimisation efforts. This article explores what these errors mean, why they occur, how they impact users, and what workarounds are available.
What Are Lighthouse Throttling Errors?
When Google’s testing infrastructure is momentarily overwhelmed by the volume of audit requests, throttling errors in Lighthouse happen. These issues specifically pertain to Lighthouse’s cloud-based component, which Google calls “Lightrider.”
Lightrider is a headless Chrome variant that performs performance audits and renders webpages. Lightrider may reject or postpone new audit requests if it receives too many at once, leading to mistakes such as:
extensible_stubs::UNABLE_TO_RETRYTHROTTLED_TASK_LIMITToo many render requestsRenderStreamService.RenderStream throttled
These messages essentially tell the user that Lighthouse attempted to audit the webpage but was unable to complete the process due to server limitations.
Why Do These Errors Occur?
Lighthouse errors of this nature are typically caused by one or more of the following factors:
- High traffic on Google’s auditing infrastructure: During peak times, many users might be running audits simultaneously, which strains the servers.
- Internal limits on parallel tasks: Google applies throttle limits to prevent individual users or systems from overloading the service.
- Memory constraints in the audit environment: If a webpage is particularly complex or heavy, it may exceed the memory limits of the audit system, especially when multiple instances are already running.
- Retry logic being disabled for efficiency: Lighthouse does not always attempt retries when system-level errors are encountered, leading to immediate failure.
These issues are not directly related to the website being tested but are instead rooted in the architecture of the Lighthouse service itself.
Examples of the Error
The following error messages are common and may appear during a PageSpeed Insights audit:
Oops! Something went wrong. extensible_stubs::UNABLE_TO_RETRY: A retriable error could not be retried due to Extensible Stubs memory limits for streams. Original error: throttling::THROTTLED_TASK_LIMIT. Another example includes: chrome-lightrider-lightrider throttled: Task is overloaded (RenderServerLoadTracker: Too many render requests).
These messages indicate that the request was refused by the server due to task or resource limits and was not retried.
Use Cases Affected by This Error
These Lighthouse errors can affect several workflows:
- SEO audits for site speed and Core Web Vitals
- Technical site performance evaluations as part of development or QA testing
- Client reports where third-party verification of site speed is required
- Benchmarking site speed before or after deployments
When such errors occur, these tasks are delayed or disrupted, potentially impacting timelines and deliverables.
Benefits of Using Google Lighthouse
Despite these occasional issues, Lighthouse remains one of the most comprehensive and accessible performance testing tools available. Its benefits include:
- Integrated with Chrome DevTools for local testing
- Part of Google’s Core Web Vitals ecosystem, making it relevant for search rankings
- Actionable suggestions for improving site performance and accessibility
- Support for both desktop and mobile audits
- Open-source nature, allowing developers to customise tests
These benefits make Lighthouse an industry standard, especially for teams focused on performance-driven development.
Challenges Associated with Lighthouse Throttling
When Lighthouse encounters throttling issues, users face a number of challenges:
1. Delayed Testing
Automated audits may be postponed due to server limitations, delaying crucial feedback loops for developers or marketers.
2. Inconsistent Results
If a test partially completes or fails during rendering, results can be inconsistent or misleading, especially for stakeholders unfamiliar with technical errors.
3. Blocked Automation
Teams using Lighthouse in CI/CD pipelines or automated tools may experience failures in scheduled tasks.
4. Lack of Retry
Unlike other systems that implement retry logic for transient errors, Lighthouse often fails silently or with a hard stop, leaving users to re-run tests manually.
5. No Official Workaround from Google
Currently, there is no formal fix or configuration to prevent this error from occurring. Google’s infrastructure manages load balancing internally.
Recommended Workarounds
While you cannot fix this issue from your end, several workarounds can reduce its impact:
Use Lighthouse Locally
- Open your site in Chrome.
- Right-click and select Inspect.
- Go to the Lighthouse tab.
- Run the test directly from your machine.
This bypasses Google’s cloud infrastructure and provides immediate results.
Try at Off-Peak Times
Running tests early in the morning or late in the evening (relative to global traffic patterns) may reduce the likelihood of hitting server limits.
Use Alternative Tools
If Lighthouse is unavailable, other tools can fill the gap temporarily:
- GTmetrix: Offers detailed performance reports using Lighthouse and Web Vitals data.
- WebPageTest.org: Allows advanced testing from multiple locations and devices.
- Pingdom Tools: Good for quick uptime and speed checks.
Monitor Google’s Status Pages
Though rare, systemic outages or degraded performance may be reported via Google’s status dashboards or developer forums.
Conclusion and Outlook
Although annoying, lighthouse throttling errors do not indicate the health of your website. They frequently happen during times of high demand and are caused by restrictions in Google’s cloud-based auditing environment. You can avoid needless troubleshooting on your own systems by being aware of the nature of these errors.
The need for tools like Lighthouse will only increase as web development continues to place a greater emphasis on performance optimisation. Even though Google hasn’t made any official announcements regarding Lightrider’s error handling, these issues should eventually be resolved by continuing infrastructure improvements.
In the interim, users are advised to maintain a consistent performance testing workflow by adding local audits and other tools to Lighthouse.
