<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://bugs.webkit.org/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.4.1"
          urlbase="https://bugs.webkit.org/"
          
          maintainer="admin@webkit.org"
>

    <bug>
          <bug_id>170074</bug_id>
          
          <creation_ts>2017-03-24 13:39:55 -0700</creation_ts>
          <short_desc>Re-enable the network and web processes&apos; keychain access to fix client certificate authentication</short_desc>
          <delta_ts>2017-03-25 14:20:00 -0700</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>WebKit</product>
          <component>WebKit2</component>
          <version>WebKit Nightly Build</version>
          <rep_platform>Unspecified</rep_platform>
          <op_sys>Unspecified</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          <see_also>https://bugs.webkit.org/show_bug.cgi?id=163710</see_also>
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords>InRadar</keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="John Wilander">wilander</reporter>
          <assigned_to name="John Wilander">wilander</assigned_to>
          <cc>achristensen</cc>
    
    <cc>bfulgham</cc>
    
    <cc>commit-queue</cc>
    
    <cc>webkit-bug-importer</cc>
    
    <cc>wilander</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>1290977</commentid>
    <comment_count>0</comment_count>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-24 13:39:55 -0700</bug_when>
    <thetext>https://bugs.webkit.org/show_bug.cgi?id=163710 caused a regression in client certificate authentication for macOS.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1290978</commentid>
    <comment_count>1</comment_count>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-24 13:41:38 -0700</bug_when>
    <thetext>rdar://problem/31095987</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1290983</commentid>
    <comment_count>2</comment_count>
      <attachid>305321</attachid>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-24 13:46:19 -0700</bug_when>
    <thetext>Created attachment 305321
Patch</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291069</commentid>
    <comment_count>3</comment_count>
      <attachid>305321</attachid>
    <who name="Brent Fulgham">bfulgham</who>
    <bug_when>2017-03-24 16:40:40 -0700</bug_when>
    <thetext>Comment on attachment 305321
Patch

This looks great to me. r=me.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291070</commentid>
    <comment_count>4</comment_count>
      <attachid>305321</attachid>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-24 16:42:44 -0700</bug_when>
    <thetext>Comment on attachment 305321
Patch

Thanks, Brent!</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291077</commentid>
    <comment_count>5</comment_count>
      <attachid>305321</attachid>
    <who name="Alexey Proskuryakov">ap</who>
    <bug_when>2017-03-24 16:48:51 -0700</bug_when>
    <thetext>Comment on attachment 305321
Patch

Shouldn&apos;t we remove the ifdef from WebProcess.sb too? There are still loads that occur in WebContent.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291084</commentid>
    <comment_count>6</comment_count>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-24 16:57:30 -0700</bug_when>
    <thetext>I do not have a test case that requires a WebProcess.sb changes to pass. My failing test only needed the change in this patch. I&apos;m happy to revisit.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291087</commentid>
    <comment_count>7</comment_count>
    <who name="Brent Fulgham">bfulgham</who>
    <bug_when>2017-03-24 17:03:12 -0700</bug_when>
    <thetext>(In reply to Alexey Proskuryakov from comment #5)
&gt; Comment on attachment 305321 [details]
&gt; Patch
&gt; 
&gt; Shouldn&apos;t we remove the ifdef from WebProcess.sb too? There are still loads
&gt; that occur in WebContent.

We believe the loads in WebContent are doing the right thing using the updated APIs, and so we want to keep them.

If we discover evidence that we need to weaken WebProcess, too, we can do so once we have data indicating it is actually needed.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291091</commentid>
    <comment_count>8</comment_count>
      <attachid>305321</attachid>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2017-03-24 17:12:58 -0700</bug_when>
    <thetext>Comment on attachment 305321
Patch

Clearing flags on attachment: 305321

Committed r214389: &lt;http://trac.webkit.org/changeset/214389&gt;</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291092</commentid>
    <comment_count>9</comment_count>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2017-03-24 17:13:02 -0700</bug_when>
    <thetext>All reviewed patches have been landed.  Closing bug.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291119</commentid>
    <comment_count>10</comment_count>
    <who name="Alexey Proskuryakov">ap</who>
    <bug_when>2017-03-24 19:07:44 -0700</bug_when>
    <thetext>There was no evidence that the original change was safe, and I don&apos;t see any evidence presented that it&apos;s safe to keep in WebProcess.

What did you test to see client certificate authentication work in WebProcess?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291120</commentid>
    <comment_count>11</comment_count>
    <who name="Alexey Proskuryakov">ap</who>
    <bug_when>2017-03-24 19:11:21 -0700</bug_when>
    <thetext>Resource loading has always required identical sandbox rules in WebContent and Networking processes, so it&apos;s quite unlikely that this case is different.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291130</commentid>
    <comment_count>12</comment_count>
    <who name="Alexey Proskuryakov">ap</who>
    <bug_when>2017-03-24 19:45:02 -0700</bug_when>
    <thetext>The reason why I&apos;m re-opening this bug and not just filing a new one is that I do believe that the above invariant is important, so we should either restore it right away, or positively prove that it&apos;s not required any more.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291154</commentid>
    <comment_count>13</comment_count>
    <who name="Brent Fulgham">bfulgham</who>
    <bug_when>2017-03-24 22:40:50 -0700</bug_when>
    <thetext>That is certainly important information. It&apos;s a shame no one on the architecture team ever bothered to document this anywhere!

Can you share which aspects of loading must be kept the same between the two processes? Certainly there are many more exceptions in WebProcess than NetworkProcess. Are any of them running afoul of this invariant,too?

John: Can you please provide a supplemental patch to support this undocumented invariant, and please add a comment somewhere so that we remember to keep these in sync in the future?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291218</commentid>
    <comment_count>14</comment_count>
      <attachid>305382</attachid>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-25 10:31:04 -0700</bug_when>
    <thetext>Created attachment 305382
Patch</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291219</commentid>
    <comment_count>15</comment_count>
    <who name="John Wilander">wilander</who>
    <bug_when>2017-03-25 10:33:32 -0700</bug_when>
    <thetext>My patch reverts the version check instead of doing an exact match with the network process sandbox. I don&apos;t want to change pre-existing sandbox behavior before we get client certificate authentication working again.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291251</commentid>
    <comment_count>16</comment_count>
      <attachid>305382</attachid>
    <who name="Brent Fulgham">bfulgham</who>
    <bug_when>2017-03-25 13:21:32 -0700</bug_when>
    <thetext>Comment on attachment 305382
Patch

R=me</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291257</commentid>
    <comment_count>17</comment_count>
      <attachid>305382</attachid>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2017-03-25 14:19:57 -0700</bug_when>
    <thetext>Comment on attachment 305382
Patch

Clearing flags on attachment: 305382

Committed r214404: &lt;http://trac.webkit.org/changeset/214404&gt;</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1291258</commentid>
    <comment_count>18</comment_count>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2017-03-25 14:20:00 -0700</bug_when>
    <thetext>All reviewed patches have been landed.  Closing bug.</thetext>
  </long_desc>
      
          <attachment
              isobsolete="1"
              ispatch="1"
              isprivate="0"
          >
            <attachid>305321</attachid>
            <date>2017-03-24 13:46:19 -0700</date>
            <delta_ts>2017-03-25 10:31:01 -0700</delta_ts>
            <desc>Patch</desc>
            <filename>bug-170074-20170324134618.patch</filename>
            <type>text/plain</type>
            <size>1916</size>
            <attacher name="John Wilander">wilander</attacher>
            
              <data encoding="base64">SW5kZXg6IFNvdXJjZS9XZWJLaXQyL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09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</data>

          </attachment>
          <attachment
              isobsolete="0"
              ispatch="1"
              isprivate="0"
          >
            <attachid>305382</attachid>
            <date>2017-03-25 10:31:04 -0700</date>
            <delta_ts>2017-03-25 14:19:57 -0700</delta_ts>
            <desc>Patch</desc>
            <filename>bug-170074-20170325103104.patch</filename>
            <type>text/plain</type>
            <size>2215</size>
            <attacher name="John Wilander">wilander</attacher>
            
              <data encoding="base64">SW5kZXg6IFNvdXJjZS9XZWJLaXQyL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09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==
</data>

          </attachment>
      

    </bug>

</bugzilla>