Minidump

Sentry can process Minidump crash reports, a memory dump used on Windows and by open-source libraries like Breakpad or Crashpad. You can either choose to generate and upload minidumps yourself or use a higher-level SDK for platforms with built-in support for native crashes:

In order to receive symbolicated stack traces, you have to upload debug information to Sentry. For more information, see Uploading Debug Information.

Platform and Language Support

Minidumps are currently supported for Windows, macOS and Linux. There is no limitation as to which programming language can be used. Sentry can also demangle symbols from the following languages; other languages will show the mangled name instead:

  • C and C++
  • ObjectiveC and ObjectiveC++
  • Swift
  • Rust

What is a Minidump?

Minidumps are files containing the most important memory regions of a crashed process. When the process crashes, the minidump is written to the user’s disk and can later be uploaded to Sentry. A minidump typically includes:

  • The runtime stack of each thread that was active during the time of the crash. This allows to reconstruct stack traces for all stacks and even infer variable values in some cases.
  • Thread contexts, i.e. register values, at the time of the crash. This is especially relevant for stackwalking.
  • Optionally, the process heap. By default, this is not included in order to keep minidumps at a reasonable size. Sentry does not read the heap, so it can be safely omitted.
  • The crash reason and an optional memory address associated to it, e.g. for memory access violations. In case of assertions, the assertion message is also included in the dump.
  • Meta data about the CPU architecture and the user’s operating system.

In addition to this information, you can add further meta data specific to Sentry, which can help in organizing and analyzing issues. For more information, see Passing Additional Data.

Creating and Uploading Minidumps

Depending on your operating system and programming language, there are various alternatives to create minidumps and upload them to Sentry. See the following resources for libraries that support generating minidump crash reports:

If you have already integrated a library that generates minidumps and would just like to upload them to Sentry, you need to configure the Minidump Endpoint URL, which can be found at Project Settings > Client Keys (DSN). This endpoint expects a POST request with the minidump in the upload_file_minidump field:

$ curl -X POST \
  '___MINIDUMP_URL___' \
  -F upload_file_minidump=@mini.dmp

To send additional information, simply add more form fields to this request. For a full description of fields accepted by Sentry, see Passing Additional Data.

Passing Additional Data

You can add more information to crash reports simply by adding more fields to the upload HTTP request. All these fields will be collected in the “Extra Data” section in Sentry:

$ curl -X POST \
  '___MINIDUMP_URL___' \
  -F upload_file_minidump=@mini.dmp \
  -F custom_field=value

Additionally, you can set all attributes corresponding to the Sentry event interface in a sentry field. This field either accepts JSON data or its values can be flattened with the bracket syntax. For example, to set the release and add a tag, send:

# As JSON
$ curl -X POST \
  '___MINIDUMP_URL___' \
  -F upload_file_minidump=@mini.dmp \
  -F 'sentry={"release":"1.2.3","tags":{"mytag":"value"}}'

# flattened
$ curl -X POST \
  '___MINIDUMP_URL___' \
  -F upload_file_minidump=@mini.dmp \
  -F 'sentry[release]=1.0.0' \
  -F 'sentry[tags][mytag]=value'

For the full list of supported attributes, see Attributes and linked documents.

Event Attachments (Preview)

Besides the minidump file, Sentry can optionally store additional files uploaded in the same request, such as log files. Simply add more files to the multipart form body. Note that the entire request must not exceed 20MB in size. Sentry will use the provided file names and mime types and list those files in the Event Attachments section at the bottom of the Issue Details page:

$ curl -X POST \
  '___MINIDUMP_URL___' \
  -F upload_file_minidump=@mini.dmp \
  -F some_file=@some_file.txt \
  -F db_log=@db.log

The organization and project setting Store Native Crash Reports also enables storage of the original minidump files. For data privacy reasons, this setting is by default disabled. Raw minidumps are deleted permanently with their issues or after 30 days.

Uploading Debug Information

To allow Sentry to fully process native crashes and provide you with symbolicated stack traces, you need to upload Debug Information Files (sometimes also referred to as Debug Symbols or just Symbols). We recommend to upload debug information during your build or release process.

Note that you need to provide debug information for all libraries that you would like to receive symbolication for. This includes dependencies and operating system libraries. If you are not sure which files are required, go to Project Settings > Processing Issues, which shows a list of all required files and instructions to retrieve them.

In addition to Debug Information Files, Sentry needs Call Frame Information (CFI) to extract accurate stack traces from minidumps of optimized release builds. CFI is usually part of the executables and not copied to debug symbols. Unless you are uploading Breakpad symbols, be sure to also include the binaries when uploading files to Sentry.

For more information on uploading debug information and their supported formats, see Debug Information Files.