1028178 - Test failure 'Modal dialog has been found and processed' in /testSecurity/testMD5HashSignature.js

Status: RESOLVED FIXED

(Core Graveyard :: Security: UI, defect, P1)

Description

[Andrei Eftimie]

Module:    testMD5HashSignature
Test:      /testSecurity/testMD5HashSignature.js    
Failure:   Modal dialog has been found and processed
Branches:  default
Platforms: All
Sample Report: http://mozmill-daily.blargon7.com/#/remote/report/74dbcf1bcfcf95d7600c6041dd375bd8

Comment 1

[Andrei Eftimie]

Attachment: skip.patch

Comment 2

[Andrei Eftimie]

Disabled:
http://hg.mozilla.org/qa/mozmill-tests/rev/78c63688f7ed (default)

Comment 3

[Andrei Eftimie]

Code fails when opening http://mozqa.com/data/firefox/security/certificates/md5/importSSL.php
> <?php
> 
> header("Content-Type: application/x-x509-ca-cert");
> include("./ssl-md5-mozqa-ca.pem");
> 
> ?>
Which only tries to include the cert at: http://mozqa.com/data/firefox/security/certificates/md5/ssl-md5-mozqa-ca.pem

Pushlog: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=bff872c9d4b2&tochange=9bd1ca579412

Regressor looks to be bug 611752

Brian, is this a genuine regression?
If not do we have to change way we import this certificate? If this is the case please advice.

Comment 4

[Dana Keeler (she/her) (use needinfo) [:keeler] (on leave)]

Turns out this shouldn't have been removed:

https://hg.mozilla.org/mozilla-central/diff/90b838762663/security/manager/pki/src/nsNSSDialogs.cpp

(although, incidentally, see bug 1024871 regarding importing a CA over http, which we probably don't want to actually allow)

Comment 5

[Dana Keeler (she/her) (use needinfo) [:keeler] (on leave)]

Attachment: patch

Comment 6

[Dana Keeler (she/her) (use needinfo) [:keeler] (on leave)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/813001854b74

Comment 7

[Dana Keeler (she/her) (use needinfo) [:keeler] (on leave)]

Marking leave-open until the mozmill test gets re-enabled.

Comment 9

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Thanks for the quick fix. In the future it would be good to spawn a new bug for the actual fix in Firefox itself, so that we do not group two different kinds of products. Given that the fix for Firefox in here is more important, I will move this bug over to security.

Comment 10

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/813001854b74

Comment 12

[Andrei Eftimie]

Reenabled the test by backing out the skip commit:
http://hg.mozilla.org/qa/mozmill-tests/rev/cdbbcca419fc (default)

I tested this with a recent mc tinderbox build and all works fine.