--- title: "Crash reporting and error analytics for mobile and web" description: "Crash groups, readable stacks and crash-free rates by version, next to the behavior that led there." url: https://countly.com/feature/crashes-errors last_modified: 2026-11-11 --- Crashes & Errors # Find the release that broke it. Switch it off without another Crash groups, readable stacks and crash-free rates by version, next to the behavior that led there. When the crashing code sits behind an Adaptive Variable, turn it off from Countly instead of shipping a hotfix. In this demo connected-car app, a release broke charging. This is the crash group behind it, with the stack that names the code. Part of analytics ## A bad release, from first crash to recovery One story from the demo app: a release breaks charging, the crash points at a flagged feature, and turning it off brings the numbers back. - Find the crash Start from what changed, find the crash group behind it, and read how many profiles and sessions it hit. - Read it, and file it The stack names the code. File it as a Jira task without leaving the page. - Turn the feature off The crashing code sits behind an Adaptive Variable. Switch it off for everyone, without a release. - Watch it recover Set an alert on completed charges so you hear the day they drop again, then mark the group resolved and watch the crash list move. If it crashes again in a newer version, it comes back as regressed. After a rollout ## The questions engineers ask the morning after A release is judged by comparison, not by one number. Each of these has its answer on a screen of its own. “Did this release make it worse?” Crash rate by version Crashes per 1k sessions for every version, next to the one before it, so a busy release and a quiet one compare fairly. “What broke overnight?” Issues, by release Filter to the new version, sort by occurrences, and see how much of it terminated the app. “Why does it crash?” The crash group A readable stack, the device’s memory and state, and the paths profiles walked right before it crashed. “Didn’t we fix that?” Regressed A group resolved in one version comes back as regressed when it crashes again in a newer one. Into the engineering workflow ## Readable stacks, filed where engineers work A crash is only useful if someone can read it and someone owns it. Symbolication ### Function names, not addresses Upload the symbol files from your build, and release crashes turn back into function names, files and line numbers. Android · Android native · iOS · macOS · tvOS · watchOS · JavaScript Jira ### One issue per crash, kept in step Create a Jira issue from a crash, with a link back to it. Resolve it on either side, and the other follows. Create issue · Link back · Two-way status Hooks ### The first crash, wherever you work A crash the project has never reported before can post to Slack, call your on-call tool or email a team. New crash trigger · HTTP · Email Works with ## Crashes & Errors, with the rest of Countly Every feature reads the same live data, so an answer in Crashes & Errors doesn’t stop there. These are the handoffs built in. [All features](https://countly.com/platform/features) Adaptive Variables Switch off the feature behind a crash for everyone, without shipping a release. Open [(Adaptive Variables Switch off the feature behind a crash for everyone, without ](https://countly.com/feature/adaptive-variables) Alerts Watch crashes as an alert source, and get told when they cross a threshold before anyone opens the page. Open [(Alerts Watch crashes as an alert source, and get told when they cross a thresho](https://countly.com/feature/alerts) Drill Every crash report is an event, so Drill can break crashes down like anything else. Open [(Drill Every crash report is an event, so Drill can break crashes down like anyt](https://countly.com/feature/drill) Journeys A crash can start a journey: apologize, or point to the fixed version. Open [(Journeys A crash can start a journey: apologize, or point to the fixed version.](https://countly.com/feature/journeys) Profiles Open one occurrence and see the profile behind it, with its own stack and device facts. Open [(Profiles Open one occurrence and see the profile behind it, with its own stack ](https://countly.com/feature/profiles) Hooks A crash the project has never seen before can start a hook that posts it wherever your team works. Open [(Hooks A crash the project has never seen before can start a hook that posts it ](https://countly.com/feature/hooks) SDK Manager Turn crash tracking on or off from the server, and set how many breadcrumbs and trace lines a crash may carry. Open [(SDK Manager Turn crash tracking on or off from the server, and set how many bre](https://countly.com/feature/sdk-manager) Data ownership ## Stack traces stay in your environment A crash report carries the stack, device facts, app state and the profile behind it. It stays on the server you chose, with the rest of your data. [Data privacy](https://countly.com/platform/data-privacy) Aspect Flex Enterprise Environment Dedicated single-tenant instance Your cloud, your data center or air-gapped Data residency The Google Cloud region you choose Wherever you deploy, including air-gapped networks Vendor access Operated by Countly; access limited, granted by you and logged No default access ISO 27001 Certified ISO 27017 Certified ISO 27018 Certified SOC 2 Type II examined TISAX® Participant* *Countly Ltd. is a TISAX® Participant. TISAX® is a registered trademark of and governed by ENX Association. Learn more at [enx.com/tisax](https://enx.com/tisax)