Expand mixed-content download protection to all http downloads
Categories
(Core :: DOM: Security, enhancement, P5)
Tracking
()
People
(Reporter: ckerschb, Unassigned)
References
(Depends on 1 open bug, Blocks 1 open bug)
Details
(Keywords: parity-chrome, Whiteboard: [domsecurity-backlog1], [wptsync upstream])
Attachments
(1 file, 1 obsolete file)
2.87 KB,
text/plain
|
chutten
:
data-review+
|
Details |
Within Bug 1614969 we started to block insecure (http) downloads on secure (https) pages. We should consider to expand our protection mechanism and start to warn generally for all http
downloads, not only mixed content downloads.
Since we have the overrule mechanism
which allows an end user to overrule the decision of Firefox, I suppose we just
have to remove the mixed-content pre-condition.
Reporter | ||
Updated•10 months ago
|
Reporter | ||
Updated•10 months ago
|
Updated•10 months ago
|
Updated•10 months ago
|
Reporter | ||
Comment 1•10 months ago
|
||
Reporter | ||
Comment 2•10 months ago
|
||
FWIW, I filed a content review request (https://mozilla-hub.atlassian.net/browse/UXREQ-171) to get some support for the two strings we are updating.
Reporter | ||
Comment 3•9 months ago
|
||
I have updated the patch to re purpose the pref dom.block_download_insecure
. Here is the TRY run:
https://treeherder.mozilla.org/jobs?repo=try&revision=fa4955d59619a63ffc6bf1fc65522aa09c719d2b
Reporter | ||
Comment 4•9 months ago
|
||
Reporter | ||
Comment 5•9 months ago
|
||
Comment 6•9 months ago
|
||
Comment on attachment 9381008 [details]
data_review_request.txt
DATA COLLECTION REVIEW RESPONSE:
Is there or will there be documentation that describes the schema for the ultimate data set available publicly, complete and accurate?
Yes.
Is there a control mechanism that allows the user to turn the data collection on and off?
Yes. This collection can be controlled through Firefox's Preferences.
If the request is for permanent data collection, is there someone who will monitor the data over time?
No. This collection will expire in six months.
Using the category system of data types on the Mozilla wiki, what collection type of data do the requested measurements fall under?
Category 1, Technical.
Is the data collection request for default-on or default-off?
Default on for all channels.
Does the instrumentation include the addition of any new identifiers?
No.
Is the data collection covered by the existing Firefox privacy notice?
Yes.
Does the data collection use a third-party collection tool?
No.
Result: datareview+
Comment 8•9 months ago
|
||
Backed out for causing wpt failures on iframe_sandbox_navigation_download_allow_downloads.sub.tentative.html.
Reporter | ||
Comment 9•9 months ago
|
||
Freddy and I agreed to rewrite the wpt-test name to include https
so the test runs relying on https rather than http.
In particular we renamed
iframe_sandbox_navigation_download_allow_downloads.sub.tentative.html
to
iframe_sandbox_navigation_download_allow_downloads.sub.tentative.https.html
Comment 10•9 months ago
|
||
Reporter | ||
Comment 12•9 months ago
•
|
||
Release Note Request (optional, but appreciated)
[Why is this notable]:
We are expanding our download protection. Previously we blocked mixed content downloads. More precisely we blocked http
downloads in https
pages. Now we are expanding our download protection to block all downloads that do not rely on a potentially trustworthy URI. In practice and in most cases this translates to http
downloads. FWIW, we initially block the load though offer the end user an option to overrule the decision of Firefox and allow the download to proceed.
[Suggested wording]:
Firefox expands download protection and initially blocks downloads relying URLs that are considered non potentially trustworthy.
[Links (documentation, blog post, etc)]:
We have and old blogpost, from the previous download protection: https://blog.mozilla.org/security/2021/10/05/firefox-93-protects-against-insecure-downloads/
Comment 13•9 months ago
|
||
Backed out for causing wpt failures @ the-iframe-element/iframe_sandbox_window_open_download_allow_downloads.tentative.html
Backout link: https://hg.mozilla.org/integration/autoland/rev/c90923a84670e17105b275c5be4a3c2842cb248e
Reporter | ||
Comment 15•9 months ago
|
||
(In reply to Sandor Molnar[:smolnar] from comment #13)
Backed out for causing wpt failures @ the-iframe-element/iframe_sandbox_window_open_download_allow_downloads.tentative.html
Similar to comment 8, I fixed the problem by letting run the WPT test in question over https
. In particular, I renamed
frame_sandbox_window_open_download_allow_downloads.tentative.html
to
frame_sandbox_window_open_download_allow_downloads.tentative.https.html
Reporter | ||
Updated•9 months ago
|
Comment 16•9 months ago
|
||
Comment 17•9 months ago
|
||
bugherder |
Comment 19•9 months ago
|
||
Added to the Fx125 relnotes with some wording tweaks.
Firefox has expanded its download protection and now more proactively blocks downloads from URLs that are considered to be potentially untrustworthy.
Updated•7 months ago
|
Comment 22•7 months ago
|
||
Backed out from autoland for breaking HTTP downloads:
https://hg.mozilla.org/integration/autoland/rev/6eb683460ad916c19b9925c30623c189c97e1f96
Comment 23•7 months ago
|
||
Backed out from mozilla-beta for breaking HTTP downloads:
https://hg.mozilla.org/releases/mozilla-beta/rev/86264e764b29445c4ae1d6b94b5354bd6d60f21e
Comment 24•7 months ago
|
||
And backed out from Release for Desktop 125.0.2 / Android 125.2.0 also.
https://hg.mozilla.org/releases/mozilla-release/rev/3840c3f2730f33a33f4325c30eddcf53ad23a48d
Comment 25•7 months ago
|
||
Backout merged to central : https://hg.mozilla.org/mozilla-central/rev/6eb683460ad916c19b9925c30623c189c97e1f96
Reporter | ||
Comment 27•7 months ago
|
||
Thanks Aryx for backing it out. I guess we need to resolve the general download machinery issues and then we can tackle this bug again. Though I am on it.
Comment 28•7 months ago
|
||
This should also go through more extensive QA when it re-lands. Please reach out to Tania or Tracy if you need help navigating the process for requesting that.
Reporter | ||
Comment 30•7 months ago
|
||
Marking this Bug as dependent of Bug 1892069, because only once we have fixed our download manager we can-reconsider landing this Bug.
FWIW, we also need to ensure we integrate the RFC1918 exception mentioned within Bug 1892011 before re-landing.
Reporter | ||
Comment 31•7 months ago
|
||
After re-discussing this feature during our HTTPS Adoption Strategy
we decided to de-prioritize this download security enhancement and focus on other HTTPS adoption features.
In case we pick this up again, we should ensure:
(a) The POST/GET download problems form Bug 1892069 are resolved.
(b) We incorporate the RFC1918 carve-outs as discussed within Bug 1892011
(c) Eliminate user annoyance. E.g. remember the triggering origin when a user clicks 'allow download'. We could store that such information in the permission manager. Whenever a user then downloads something over HTTP from that same website (triggering origin/triggeringPrincipal), then the download would be allowed without Firefox interfering/blocking the download initially and having the end user to click the allow download exception.
Updated•7 months ago
|
Description
•