CITGM fail

[refack]

Maybe just a load induced timeout...

 274 passing (5s)
   1 failing
   1) WebSocket permessage-deflate #terminate will raise error callback, if any, if called during send data:
      Error: Timeout of 2000ms exceeded. For async tests and hooks, ensure "done()" is called; if returning a Promise, ensure it resolves.

Same error on osx1010 and aix61-ppc64
Test: https://github.com/websockets/ws/blob/master/test/WebSocket.test.js#L1990
CITGM report: https://ci.nodejs.org/view/Node.js-citgm/job/citgm-smoker/813/testReport/

[refack]

This is the diff since last passing nodejs/node@171a43a...ad4765a
Prbly nodejs/node@330c8d7

[mcollina]

@refack I do not think opening a bug on the target module is the best approach for core. You should really open a bug in core.

[refack]

@refack I do not think opening a bug on the target module is the best approach for core. You should really open a bug in core.

I usually do, initially I thought it was a load induced timeout.
Now @lpinca seems to agree nodejs/node#12828 (comment)

P.S. It's CITGM policy to inform module authors https://github.com/nodejs/citgm/blob/master/CONTRIBUTING.md#submitting-a-module-to-citgm

[mcollina]

Anyway, as the author of the offending issue I would have loved to know :).
I'm a bit unsure how I missed that, and why it fails only on AIX and OSX, but not everywhere else.

[refack]

Anyway, as the author of the offending issue I would have loved to know :).
I'm a bit unsure how I missed that, and why it fails only on AIX and OSX, but not everywhere else.

Big Ack 🤦‍♂️
Re: nodejs/node#12925 (comment)

[mcollina]

Anyway, it's confirmed and a fix is on its way.

[lpinca]

Awesome, thanks.