[Webkit-unassigned] [Bug 208049] Javascript can't access a SameSite=Strict cookie after page is loaded after a redirect from a third party site

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Dec 8 10:01:22 PST 2020


https://bugs.webkit.org/show_bug.cgi?id=208049

Peleg Rosenthal <peleg3 at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |peleg3 at gmail.com

--- Comment #14 from Peleg Rosenthal <peleg3 at gmail.com> ---
We're seeing the same issue in Safari 14.01: same-site cookies set after a redirect from a third-party site aren't accessible via `document.cookie`.

Similarly to OP, we rely on double-submit CSRF cookies. We noticed the issue when Sendgrid's click-tracking redirects to our site and our cookies are no longer accessible to JS.

A few other things we've noticed:

1. Issue persists after refreshing the page. For example: Sendgrid redirects to first-party (ie xyz.com); cookie inaccessible to JS; refresh xyz.com; `Set-Cookie` header is in the response; cookie is included in subsequent xhr requests, yet is still inaccessible to JS.

2. Pre-existing cookies that were accessible to JS, after a redirect, are no longer accessible to JS. For example: cookie accessible to JS w/in xyz.com, but when user lands on xyz.com after a redirect, the same cookie is no longer accessible to JS.

3. In addition to not being accessible via `document.cookie`, the aforementioned cookies are also not shown in the "Storage" tab in the web inspector.

In the meantime we've mitigated this issue by reducing same-site level to "Lax," which works given we don't use CSRF protection for GETs, but it'd be ideal if Strict cookies could be used in the same way.

This was all tested on Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_6) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.0.1 Safari/605.1.15

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20201208/0387e72e/attachment.htm>


More information about the webkit-unassigned mailing list