You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We've noticed that if we send SIGKILL to an Nx process sometimes it can crash from within the Rust Nx code and hang indefinitely. This issue is much more prevalent on our private monorepo which has 709 projects, but I've been able to create a reproduction case https://github.com/JamesNimlos/nx-debug-on-sigkill.
Expected Behavior
Nx should not panic reopen a terminal session and honestly I'm not entirely sure how it could since it's a SIGKILL.
Again, we see this very consistently on our private monorepo which has 709 and millions of lines of TS which obviously means it will take longer resolving the graph etc.
Additionally, on the reproduction case I could only get it to reproduce when I use the start target and added the build target to dependsOn.
The text was updated successfully, but these errors were encountered:
We started seeing a similar behaviour recently. Errors like:
> nx build app
thread '<unnamed>' panicked at packages/nx/src/native/pseudo_terminal/pseudo_terminal.rs:61:27:
Failed to enter raw terminal mode: Os { code: 6, kind: Uncategorized, message: "No such device or address" }
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
fatal runtime error: failed to initiate panic, error 5
Aborted (core dumped)
Current Behavior
We've noticed that if we send SIGKILL to an Nx process sometimes it can crash from within the Rust Nx code and hang indefinitely. This issue is much more prevalent on our private monorepo which has 709 projects, but I've been able to create a reproduction case https://github.com/JamesNimlos/nx-debug-on-sigkill.
Expected Behavior
Nx should not panic reopen a terminal session and honestly I'm not entirely sure how it could since it's a SIGKILL.
GitHub Repo
https://github.com/JamesNimlos/nx-debug-on-sigkill
Steps to Reproduce
The script is this:
Where it's running
nx start crew
and then 300 ms later sending that process a SIGKILL (-9
).Nx Report
Failure Logs
Package Manager Version
No response
Operating System
Additional Information
Again, we see this very consistently on our private monorepo which has 709 and millions of lines of TS which obviously means it will take longer resolving the graph etc.
Additionally, on the reproduction case I could only get it to reproduce when I use the
start
target and added thebuild
target todependsOn
.The text was updated successfully, but these errors were encountered: