Segmentation fault at address 0x0

[noahgsolomon]

How can we reproduce the crash?

Not sure the easiest way to reproduce, I'm just giving this issue because every single time for the last 2 months I've had a long winded process (30 mins to 24 hours or more), I get a segfault crash. And this is exclusively with bun. This is why I now use Deno 2 to run my long winded scripts.

So some info on how to produce this is just run a batch process that takes a while on bun pretty much.

Relevant log output

Bun v1.1.20 (ae194892) Linux x64
WSL Kernel v5.15.153 | glibc v2.35
CPU: sse42 popcnt avx avx2
Args: "bun" "run" "add-vectors-to-existing-people.ts"
Features: jsc fetch(15442) transpiler_cache tsconfig_paths tsconfig WebSocket(647) 
Builtins: "bun:main" "node:buffer" "node:crypto" "node:fs" "node:os" "node:path" "node:string_decoder" "node:util/types" 
Elapsed: 1587443ms | User: 115651ms | Sys: 69065ms
RSS: 0.02ZB | Peak: 5.98GB | Commit: 0.02ZB | Faults: 25

panic: Segmentation fault at address 0x0
oh no: Bun has crashed. This indicates a bug in Bun, not your code.

Stack Trace (bun.report)

Bun v1.1.20 (ae19489) on linux x86_64 [RunCommand]

Segmentation fault at address 0x00000000

• ld-temp.o:0: WTF::Unicode::convertReplacingInvalidSequences
• ??
• http.zig:3145: src.http.NewHTTPContext
• main.zig:64: src.deps.uws.NewSocketHandler
• openssl.c:1107: create_ssl_context_from_bun_options
• epoll_kqueue.c:380: us_loop_run_bun_tick
• array_hash_map.zig:2022: src.http.HTTPThread.processEvents
• string_immutable.zig:885: src.http.NewHTTPContext

[github-actions[bot]]

@noahgsolomon, the latest version of Bun is v1.1.29, but this crash was reported on Bun v1.1.20.

Are you able to reproduce this crash on the latest version of Bun?

bun upgrade