Bug 230076 - Regression (macOS Monterey) Websocket connection closing
Summary: Regression (macOS Monterey) Websocket connection closing
Status: RESOLVED DUPLICATE of bug 228296
Alias: None
Product: WebKit
Classification: Unclassified
Component: Page Loading (show other bugs)
Version: Safari Technology Preview
Hardware: Unspecified Other
: P2 Normal
Assignee: Nobody
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2021-09-08 18:45 PDT by isaac+webkit
Modified: 2021-09-15 00:28 PDT (History)
5 users (show)

See Also:

Attachments
Copy of the failing packet listing from the Safari Web Inspector (Monterey) (96.49 KB, text/plain)
2021-09-08 18:45 PDT, isaac+webkit 		no flags Details
Copy of successful packet listing from BigSur (109.93 KB, text/plain)
2021-09-08 18:46 PDT, isaac+webkit 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description isaac+webkit 2021-09-08 18:45:28 PDT 
Created attachment 437696 [details]
Copy of the failing packet listing from the Safari Web Inspector (Monterey)

We are experiencing unexplained Websocket connection failures/closes when connecting to our servers on macOS Monterey.

Specifically larger message payloads, after a series of smaller messages are able to reliably reproduce the issue. 
I've attached the larger messages which cause the issue.

This is not an issue on macOS Big Sur and earlier.
Comment 1 isaac+webkit 2021-09-08 18:46:02 PDT 
Created attachment 437697 [details]
Copy of successful packet listing from BigSur
Comment 2 isaac+webkit 2021-09-08 18:49:34 PDT 
If credentials etc, are needed to help diagnose this issue I can happily provide them via private mail.
Comment 3 Radar WebKit Bug Importer 2021-09-09 03:44:22 PDT 
<rdar://problem/82917968>
Comment 4 Alex Christensen 2021-09-14 08:55:48 PDT 
I reached out for those credentials.
Comment 5 Alex Christensen 2021-09-15 00:28:22 PDT 
Thank you for writing this bug and giving me access to your test setup.  It allowed me to isolate the cause of this bug and find a fix.

I isolated it to some valid UTF-8 sequences being flagged as invalid, which causes kNWErrorDomainPOSIX error 100

*** This bug has been marked as a duplicate of bug 228296 ***