React-beautiful-dnd: flow crashing on circleci

Created on 12 Dec 2018  路  17Comments  路  Source: atlassian/react-beautiful-dnd

Builds have been increasingly running into this issue:

$ flow check --max-warnings=0
Worker interrupted with signal: sigkill
Subprocess(364): fail 2Unhandled exception: Worker_failed (process_id = 364): (Worker_quit WEXITED 2)
Raised at file "src/core/lwt.ml", line 3008, characters 20-29
Called from file "src/unix/lwt_main.ml", line 42, characters 8-18
Called from file "src/commands/checkCommands.ml", line 135, characters 27-86
Called from file "src/commands/commandUtils.ml", line 13, characters 4-32

I have sshed into the box and run

yarn flow and it cannot spin up the server. It tries 7 times then exits

engineering health 馃洜

All 17 comments

I had a problem with workers in my rollup+uglify setup. Only numWorkers: 1 worked for me.

Since which version you get this error?

This consistently fails during the typecheck task (yarn flow check)

We have been getting it for a while, but this error's frequency has been increasing

It used to be 1/10 builds, now it is 8/10

@jbrown215 could you help with this issue?

@alexreardon could you try server.max_workers=1

https://flow.org/en/docs/config/options/#toc-server-max-workers-integer

Done ^

It looked like that worked @TrySound. Super strange. What do you think would be the best way to only use max_workers for ci?

I will re-run the build of #1008 to see if it works consistantly

Not sure, but some time ago i saw some custom config specifying. Will take a look tomorrow.

Thanks @TrySound

You're a hero! 馃挭

The problem is circle ci. I had the same problem with jest workers

I have merged that change for now. Ideally we only have this setting on for ci, but running local seems to be fine with max_workers=1.

I have merged for now because it fixes our builds

This looks resolved to me-- sorry I didn't get to this earlier.

All fixed

Was this page helpful?
0 / 5 - 0 ratings