Dealing with app crashes from third-party APIs. Are you prepared to address user complaints effectively?

An example I have seen is an application where a third party API was being used to deliver OTP to the user. When it was crashing, also the application on this side crashed. All we had to do is use try and catch and if that happens we instruct the user to check their email since we added sending the OTP via an email on the catch block.

As a Tech Lead, I am following below steps for our microservices architecture: ?? Implement graceful error handling and fallback options to keep the app functional during crashes. ?? Regularly monitor third-party APIs for downtime or failures and set up alerts. ?? Use retry logic and circuit breakers to manage temporary API issues. ?? Cache data to maintain functionality during API outages. ?? Communicate transparently with users about ongoing issues and fixes. ?? Push timely updates to resolve crashes and improve stability.

We gathered user feedback and created a dashboard to log both positive and negative comments. Our priority is to address the negative issues reported as quickly as possible in the upcoming sprint. We're using the MoSCoW method to prioritize tasks, categorizing them into Must-haves (critical issues that need immediate fixing), Should-haves (important but not urgent), Could-haves (nice-to-have features), and Won’t-haves (for future consideration). This approach ensures that we focus on the most critical fixes while maintaining a balanced development roadmap.

That's why it's always important to have Devops monitoring tools like ELK to read the logs and tell the cause of the problem exactly. This will help make an informed decision on what next.

As a lead developer, I have used ELK stack to monitor the application: ??Real-time alerts: Set up alerts for API failures, downtime, or crashes, allowing quick response to issues before they impact users. ??KPI metrics tracking: Monitor key metrics like API response times, error rates, and app performance to identify trends and potential problems. ??Log analysis: ELK provides detailed log data to troubleshoot crashes, pinpoint root causes, and trace issues back to specific API calls. ??Visual dashboards: Use Kibana to visualize app performance, API health, and crash patterns, making it easier to track issues over time. ??Historical insights: Analyze logs and metrics over time to prevent recurring issues and improve future API integrations.

When the app crashes, we promptly notify users and apologize for the service disruption, emphasizing our commitment to resolving the issue as quickly as possible. Transparency is key in maintaining user trust. Additionally, we are actively seeking the best solutions to prevent future failures, despite already having robust monitoring tools, an alert system, and circuit breakers in place. This proactive approach ensures that we continuously improve our systems to enhance reliability and user experience.

Actually, I agree with this. In the example I gave earlier, we had to make a mitigation strategy by using the email option to send the OTP.

In addition to viewing each crash as a learning opportunity, we also: ??Conduct post-mortem analyses: Hold regular reviews after incidents to discuss what happened, identify root causes, and derive actionable insights. ??Regularly review documentation: Stay updated with the documentation of third-party APIs to anticipate changes and ensure our application remains compatible. ??Establish a knowledge base: Create and maintain documentation of past incidents, resolutions, and best practices to aid the team in future troubleshooting efforts.