Receiving this crash consistently when using Bun workers

[chadacious]

How can we reproduce the crash?

I'll start trying to track down if it's related to any "toLowerCase()" operation in my code, but I haven't pinpointed a reproducible yet.

Relevant log output

============================================================
Bun v1.1.34 (5e5e7c60) macOS x64
macOS v13.6
CPU: sse42 popcnt avx avx2
Args: "bun" "run" "./src/index.ts"
Features: WebSocket(197) bunfig dotenv(192) fetch(15421) http_server jsc spawn(391) transpiler_cache(1566) tsconfig(10) workers_spawned(63) 
Builtins: "bun:main" "node:assert" "node:async_hooks" "node:buffer" "node:child_process" "node:constants" "node:crypto" "node:dns" "node:events" "node:fs" "node:fs/promises" "node:http" "node:https" "node:net" "node:os" "node:path" "node:process" "node:querystring" "node:stream" "node:string_decoder" "node:timers" "node:timers/promises" "node:tls" "node:tty" "node:url" "node:util" "node:util/types" "node:zlib" "node:punycode" "ws" "node-fetch" "node:http2" 
Elapsed: 3225124ms | User: 2266443ms | Sys: 775370ms
RSS: 10.07GB | Peak: 11.68GB | Commit: 0.02ZB | Faults: 482

Stack Trace (bun.report)

Bun v1.1.34 (5e5e7c6) on macos x86_64 [RunCommand]

Illegal instruction at address 0x105C3DE8C

• 2 unknown/js code
• pas_immortal_heap_allocate_with_heap_lock_hold_mode
• pas_segregated_directory_append
• pas_segregated_size_directory_take_first_eligible
• bmalloc_heap_config_specialized_local_allocator_try_allocate_small_segregated_slow
• bmalloc_allocate_impl_casual_case
• WTF::StringImpl::convertToLowercaseWithoutLocale
• WTF::String::convertToLowercaseWithoutLocale
• operationToLowerCase

Features: workers_spawned, WebSocket, bunfig, dotenv, fetch, http_server, jsc, spawn, transpiler_cache, tsconfig

_{Sentry Issue: BUN-8RW}

[github-actions]

@chadacious, the latest version of Bun is v1.1.36, but this crash was reported on Bun v1.1.34.

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

bun upgrade

[Jarred-Sumner]

Can you share any code?

[chadacious]

@chadacious, the latest version of Bun is v1.1.36, but this crash was reported on Bun v1.1.34.

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

bun upgrade

So far so good on the latest version of Bun. It would occur after a period of time and not something I pinpointed to specific code as there is no call stack and the project is pretty complex... could be anywhere. Will report back if I see it again and/or can provide a reproducible.

[chadacious]

Ok, I see that actually this is a memory leak resulting from the use of workers. Seems related to this open issue: #5709

Looking for a workaround, or some way to reset/clear memory. For now, I'll use child-process instead.

[Jarred-Sumner]

We are tracking worker stability in #15964