SIGSEGV Error appears sometimes on installing @sentry/profiling-node

[sl-ffx]

Is there an existing issue for this?

• [X] I have checked for existing issues https://github.com/getsentry/profiling-node/issues
• [X] I have reviewed the documentation https://docs.sentry.io/
• [X] I am using the latest SDK release https://github.com/getsentry/profiling-node/releases

How do you use Sentry?

Self-hosted/on-premise

SDK Version

7.93.0

Link to Sentry event

No response

What environment is your node script running in?

In a docker container using the image node:20-alpine (currently: node:20.11.0-alpine3.19) It uses node version v20.11.0 and npm version 10.2.4

How is your code deployed and bundled?

The dependencies are installed in the docker container.

Steps to Reproduce

1. Run a new container docker run --rm -it node:20-alpine sh
2. Create and change directory mkdir /bar && cd /bar
3. Initialise npm npm init -y
4. delete node_modules directory and install @sentry/profiling-node rm -rf node_modules/ && npm install --save @sentry/profiling-node
5. repeat step 4 until the error appears (sometimes you have to repeat it up to 30 times)

Expected Result

The module is installed successfully every time

Actual Result

Sometimes the installation fails with a SIGSEGV signal. On my local device, the failing rate is completely random. Sometimes it fails after 3 attempts and sometimes after 30 attempts. (docker version 24.0.7)

On my Jenkins server, it fails more often. (about every second attempt) The Jenkins is running on an "Amazon Linux 2023" instance on AWS. (docker version 24.0.5)

Here is a log of the installation: 2024-01-11T09_27_55_177Z-debug-0.log

[JonasBa]

I suspect this is the same issue as https://github.com/getsentry/profiling-node/issues/233, I'm going to create a release now with a tentative fix.

[JonasBa]

Do you mind bumping to v1.3.4 and letting me know if the issue persists?

[sl-ffx]

everything works as expected. Thank you.