Development

Crash-Proofing Your Code: A Guide to Firebase, Bugfender, and Sentry

February 7, 2024

Tomáš Dvořáček

Senior Android Developer

When you own a significant product, releasing it to the public requires more than just hope for the best. Even with developers and testers checking the product, some mistakes usually make their way into production. To identify any user-discovered issues, a monitoring system becomes crucial. These systems play a vital role in maintaining the quality and stability of software applications. While you can implement your solution, there are already numerous tools on the market designed for this purpose, so why reinvent the wheel?

Let's explore some monitoring systems:

Firebase

‍Managed by Google, this set of tools provides Crashlytics for both fatal and non-fatal crashes in your product, along with analytics and additional features. The tool is available for free; all you need is a Google email account and to configure some settings.

BugFender

This lesser-known tool handles crash logging and data collection and can also collect user feedback. In the free subscription tier, they offer a limited set of features and 24-hour log retention. For more serious projects, a paid subscription is required.

Sentry

‍Sentry is a monitoring tool offering a vast array of features. It supports various platforms, including mobile, web, and gaming frameworks, with community support for many more. Sentry tracks event logs and crashes and assists in analyzing them. It also provides insight into the performance of your product, supporting widely used programming languages. Sentry has different subscription tiers; the free tier provides basic features, which may be sufficient for solo developers. The higher the subscription, the more support and metrics your product receives. Check out the detailed features that we particulary like:

Error Tracking and Monitoring

‍Easy to add to your project, Sentry automatically collects data like device info, web browser info, crash logs, and additional user interactions. The specifics of what is tracked depend on the platform and activated Sentry plugins. For instance, in the Android system, Sentry can track user interactions, HTTP calls, visited screens, and more. This information makes it easier to identify the cause of errors.

Source: https://docs.sentry.io/product/issues/

Real-Time Alerts

‍Sentry allows you to set alerts based on various conditions. For example, if a completely new error occurs, Sentry can send you an email or trigger a webhook to receive a message in platforms like Slack or Microsoft Teams, enabling a quick response to the issue.

Source: https://docs.sentry.io/product/alerts/

Error Insights

‍Opening an issue in Sentry reveals device info, product version, details of where and what went wrong, and the actions taken by the app and user before the issue (breadcrumbs). Sentry can also provide statistics about the issue, grouped into categories, making it easier to discover patterns, such as the issue occurring only on Safari browsers or on iOS 16.

Source: https://docs.sentry.io/product/issues/issue-details/error-issues/

Supported Languages and Platforms

‍Sentry supports numerous programming languages and platforms, making it versatile for use in various products, whether in mobile development using Kotlin or Swift, or web development using React, Angular, ASP.NET, PHP, and more.

Source: https://docs.sentry.io/platforms/

Performance Metrics

‍With Sentry log collection, you can analyze the performance of specific parts of your product. This includes understanding the performance of web pages, identifying features that consume more time, or pinpointing dropped frames or lags in mobile apps.

Source: https://docs.sentry.io/product/performance/?original_referrer=https%3A%2F%2Fwww.ecosia.org%2F

Ease of Integration

‍Integrating Sentry's SDK into your product is straightforward, regardless of the platform. Sentry provides tutorials, requiring you to create an account, a new project, and add the provided ID and DNS link to your product. You can then test crashes to ensure they appear in Sentry. Additionally, many existing tools support integration with Sentry, including messaging programs like Slack or Microsoft Teams, project management tools like Jira, and git repositories like GitHub and Bitbucket.

Source: https://docs.sentry.io/?original_referrer=https%3A%2F%2Fwww.ecosia.org%2F

Self-hosting Sentry and Pricing

‍Sentry can be hosted on your server, offering more independence for companies. However, a license fee still applies.

Pros and Cons of Sentry:

Pros:

Specialized for error tracking and analysis
Supports a wide range of platforms
Offers real-time alerts
Provides detailed error insights
Supports community and open-source contribution
Many third-party integrations

‍Cons:

Not free; the price depends on the features you need
The abundance of features in Sentry can be overwhelming

This article is the first part of a two-part series. The next article will guide you on setting up Sentry for the Android platform.

Thank you for reading!

‍

What’s next?

Let’s continue the conversation.

3D representation of discovery track, glass cubes and a metal arrow that spins around them.

Stay in touch

Thank you! We will be in touch.

Oops! Something went wrong while submitting the form.

Project

1 min read

Pivot Like a Pro: Startup Strategies for Success

Cleevio

Company News

Discover the highlights from our recent exclusive panel discussion featuring Roman Stanek, CEO of GoodData, alongside industry leader Milo Poplar and our Head of North America, Filip Orth.

The event provided valuable insights into global-scale product building and entrepreneurial success. Dive into the key takeaways and memorable moments from this engaging discussion, where experienced pioneers shared their wisdom and expertise on the AI landscape.

We were thrilled to have Roman Stanek and his impressive track record in building successful startups. His journey began with the acquisition of NetBeans by Sun Microsystems in 1999, followed by the acquisition of Systinet by Hewlett Packard in 2006. Along the way, his ventures received support from renowned VC funds like a16z, Intel Capital, and General Catalyst.

Outsourcing

During the event, our discussions delved deeply into the strategies startups can employ to enhance efficiency through outsourcing across various operational domains. We explored how outsourcing can serve as a vital tool in mitigating the challenges inherent in the dynamic startup landscape. Among others, these challenges encompass navigating a tough fundraising environment and contending with fierce competition in virtually every industry.

Agility and adaptability

Central to our discussions was the imperative for startup founders to remain agile and adaptable in the face of evolving circumstances. We underscored the critical importance of being responsive to market shifts and customer demands, emphasizing the need for continual adjustment to ensure the longevity and success of their ventures.

Operational effectively

The ability to pivot effectively is essential, particularly when considering the quest for operational efficiency. In this context, leveraging outsourcing capabilities becomes paramount, whether it is in technological development, sales, marketing, or legal compliance. This strategic approach is especially pertinent in the current landscape, where fundraising endeavors are meeting with increased difficulty, and access to capital is more constrained than it was just two years ago.

Milo Poplar, Co-founder of Eldison during presentation

We intend to make this event a regular occurrence. We aim to host Founders’ Lounge regularly, each time focusing on a different theme and bringing in a special guest who will share insights from their entrepreneurial journey.

Are you a founder, investor, CTO or decision-maker in the technological startup environment seeking to accelerate your growth and gain a competitive edge in the market? Watch our LinkedIn for future invites!

Development

1 min read

Crash-Proofing Your Code: Setting Up Sentry on Android

Tomáš Dvořáček

Senior Android Developer

Welcome back! As promised, we are here with the second part of the Crash-proofing series. This time, we will show you how to set up Sentry for the Android platform and some nice-to-have stuff as a bonus.

Let's begin!

Create sentry.properties file inside your Android project’s root folder

Initial setup in Sentry Admin

These are a few things you need to do before implementing Sentry Android SDK.

Create a new project in Sentry. Give this project some appropriate name like your-project-android
Then go to Settings → Projects → Your new Project → General Settings

You should see the name of your project. This is your project_id. In sentry.properties, replace <project_id> with it
Then go to Organization settings → General Settings, copy your organisation_slug and replace <organisation_slug> inside sentry.properties

Then go to User and tokens and create new Token. You can add every permission to it for a start but its better to check every permission and decide for yourself what is needed.

Then copy your new auth token and replace <auth_token> with it inside sentry.properties.Sentry can upload ProGuard mapping files for you and needs this auth token. Ignore this if you don’t care about ProGuard
Your sentry.properties should be filled now.
Then go to Settings → Projects → Your new Project → Client Key (DSN)
Get your DSN link and save it somewhere for now.

Now you can return to your Android Project

Adding sentry to project

A full tutorial is on this page. Sentry can set itself with a command, but if you prefer a manual approach or have problems, follow the instructions below or this page. Sentry will also give you a direct link for setup in Android Studio (https://sentry.io/organizations/XXX/projects/YYY/getting-started/)

Add this to your main module build.gradle

Open your application class and initiate Sentry

Let’s explore what each line does:

This line applies the environment to the issue. If you use several build flavors like development and production, you can then easily filter specific flavors.

These lines set the percentage value of how many traces to log. 100% means everything is logged. It’s not recommended to collect everything in a production environment; however, e.g., on development, it’s a good thing since testers could find bugs. In our example, we set the percentage on the release version to 10%.

Open your AndroidManifest.xml file inside your main module and insert this

If you don’t want to use CI/CD and just hardcode your DSN, replace ${sentryDsnUrl} with your collected DSN link.

Open app’s build.gradle and setup DSN there as well

Add Sentry gradle plugin to project

This code enables sentry gradle plugin. This prepares scripts for your convenience like uploading aab to sentry with proguard mapping files.

Nice things to know

Assign someone to new issue

Every issue in Sentry can be assigned to someone and you can do that manually or automatically with rules.

Create Jira issue directly from Sentry

If you setup the Jira integration mentioned in the previous article, Sentry enables you to quickly and easily create Jira issues from the issue itself.

Discover page

Discover page let’s you quickly query in all yours collected data so you can answer analytics questions or analyse potential health issues in your project. You can use predefined queries or create, edit and delete your own queries. It’s really powerful tool but it’s limited only to Trial and Business Plan.

Pluggins for Sentry

Sentry for Android offers plugins which let’s you automatically collect more specific data. For example:

Some stuff like Android Fragments, okHttp events and logcat are already logged automatically from Sentry version 3.1.0.

In order to disable them, change the sentry config in the project’s build.gradle.

OkHttp plugin

Like said above, from Sentry 3.1.0, Sentry will automatically log OkHttp event via manipulating your final bytecode and inserting SentryOkHttpInterceptor. This will create breadcrumbs which would look like this:

‍

If you use Sentry 3.11.0+, Sentry will also insert SentryOkHttpEventListener which will log even more data from http requests. Everything will be logged with logcat:

Thank you for diving into the second part of our series. If you have any questions, feel free to catch up or reach out.

‍

#meetcleevio

1 min read

From Code to Culture: The Backbone of Our Backend Team

Přemysl Rucký

Backend Lead

Hey there! Curious about what makes our backend team tick? Let me spill the beans on what keeps us running at Cleevio.

Team Contract: Our Secret Sauce

Ever heard of a team contract? It's like our BE team's rulebook but cooler. We all chipped in to create this gem, covering everything from our core values to nifty Slack etiquettes (like no-reply expectations after 4 PM). It's not just a newbie guide; even our weathered developers learn a thing or two from it.

Bringing the Team Together

With team members scattered across Prague, Brno, and beyond borders, building bonds ain't easy, but we're not ones to back down from a challenge. That's why we came up with Backend Onsite Day. Originally a monthly affair, we scaled it back to every two months to make it work for everyone. It's a chance to catch up, brainstorm, and maybe even grab a beer together.

Sharing Know-How Like a Boss

We take knowledge sharing seriously. Our weekly BE meetings aren't just about updates and project briefs - they are also our chance to pick each other's brains. Plus, we've got knowledge base packed with tech insights and project learnings. It's like having a cheat sheet for all things backend.

Reflect, Refine, Repeat

We're all about improvement. That's why we love our team retrospectives. Just recently, at the start of 2024, we huddled up for about an hour to chat about how things were going. Turns out, everyone is pretty happy being part of the team, which is great to hear - but we also came up with some ideas to make things even better next time. It's all about celebrating wins and figuring out how we can crush it even more in the future.

Growing Together

Personal growth is big on our agenda. Regular performance reviews and one-on-one chats keep us on track. And with internal training programs, education budgets, and buddy systems in place, we're all set to level up together.

‍

At the end of the day, it all comes down to communication. Whether it's a quick Slack chat or a face-to-face meeting, keeping the lines of communication open is what keeps our team humming.

So there you have it—our backend team's recipe for success. With a healthy dose of teamwork, communication, and a sprinkle of resilience, there's no limit to what we can achieve together at Cleevio.

all posts