Closed Bug 1908161 Opened 2 years ago Closed 2 years ago

Clicking some links on Bing are not redirecting properly

Categories

(Web Compatibility :: Site Reports, defect, P1)

Firefox 128
Unspecified
Windows

Tracking

(Not tracked)

RESOLVED DUPLICATE of bug 1909666

People

(Reporter: mkaply, Unassigned)

References

()

Details

(Keywords: webcompat:needs-diagnosis, webcompat:site-report, Whiteboard: [webcompat:sightline])

User Story

platform:windows,mac,linux,android
impact:site-broken
configuration:general
affects:some
branch:release
diagnosis-team:networking

Attachments

(2 files)

If you navigate to:

https://www.bing.com/search?pc=MOZI&form=MOZLBR&q=best+prime+day+deals

and click on some of the links below the prime ad (in the screen recording, Firefox goes to a bing URL but never redirects to the actual ad.

Oddly, the resulting web page isn't actually loaded. There is no reload button. Pressing enter loads the page.

If you bring up the inspector, there is no page.

I tried to use mozregression to see if this is new and it seemed like it was? Like maybe around May 20, but I ran into other problems that kept me from properly using mozregression.

I was able to reproduce this on Windows with 128 Release and 130 Nightly FWIW

Profile:

https://share.firefox.dev/3W7QxZr

I used:

timestamp,sync,nsHttp:5,cache2:5,nsSocketTransport:5,nsHostResolver:5,EarlyHint:5,nsDocShell:5

I didn't start logging until the bing page was up.

I clicked the link and it didn't load and then I stopped logging.

Component: DOM: Navigation → Site Reports
Product: Core → Web Compatibility

Profile with more stacks

First link worked, second one didn't. Stopped log after failing link
https://share.firefox.dev/3W9aC1K

Duplicate of this bug: 1907471

I don't necessarily trust mozregression in this case because it seems intermittent, but in my testing, something definitely changed around May 20th because before May 20th, they loaded consistently.

https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=750a9d68013bf86f5d1e16ddc95cce396fffd4a2&tochange=0d8aad9a5c68f8b9bc5adcd4b7aa368bc74ddf75

Is the closest I could come. It picked some specific patches, but I don't trust it.

https://bugzilla.mozilla.org/show_bug.cgi?id=1865040 is on the list, but I don't see how it could be related.

FWIW, I can't reproduce this.

Could you try to record the http log to a file? It's easier for debugging.
Could you also try to disable network.http.priority_header.enabled and see if this still happens?

Thanks.

Flags: needinfo?(mozilla)
Flags: needinfo?(mozilla)

Still happens with that pref flipped (let me know if I would need to restart).

Also, this appears to be Windows only.

OS: Unspecified → Windows
Version: unspecified → Firefox 128

From the the profiler and moz log, I see the same thing:

2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: D/nsSocketTransport nsSocketInputStream::Read [this=20368d84dc0 count=9]
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: D/nsSocketTransport   calling PR_Read [count=9]
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: D/nsSocketTransport   PR_Read returned [n=-1]
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: D/nsSocketTransport ErrorAccordingToNSPR [in=-5961 out=804b0014]
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: D/nsSocketTransport nsSocketTransport::OnMsgInputClosed [this=2035ef24600 reason=804b0014]
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: I/nsHttp Http2Session 20360e19900 buffering frame header read failure 804b0014
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: V/nsHttp nsHttpConnection::OnSocketReadable 20363ca2400 trans->ws rv=804b0014 n=0 socketin=804b0014
2024-07-17 11:43:01.852000 UTC - [Parent 14024: Socket Thread]: V/nsHttp nsHttpConnection::CloseTransaction[this=20363ca2400 trans=20360e19900 reason=804b0014]

The connection was closed because of NS_ERROR_NET_RESET, which indicates that the server might have closed the connection. Could you also try to capture a Wireshark trace so we can confirm if we received a TCP RESET?

Thanks.

Flags: needinfo?(mozilla)

Wireshark sent separately.

Flags: needinfo?(mozilla)
Severity: -- → S2
User Story: (updated)
Priority: -- → P1
See Also: → 1909666
Status: NEW → RESOLVED
Closed: 2 years ago
Duplicate of bug: 1909666
Resolution: --- → DUPLICATE
See Also: 1909666
Whiteboard: [webcompat:sightline]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: