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

<bugzilla version="5.0.6"
          urlbase="https://bugzilla.yoctoproject.org/"
          
          maintainer="it-coreprojects-helpdesk@linuxfoundation.org"
>

    <bug>
          <bug_id>4741</bug_id>
          
          <creation_ts>2013-06-17 20:06:00 +0000</creation_ts>
          <short_desc>Netfilter NAT not configured properly/different for Kernel 3.8.4</short_desc>
          <delta_ts>2013-07-31 19:52:35 +0000</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>7</classification_id>
          <classification>Build System, Metadata &amp; Runtime</classification>
          <product>OE-Core</product>
          <component>kernel</component>
          <version>unspecified</version>
          <rep_platform>x86</rep_platform>
          <op_sys>x86</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords></keywords>
          <priority>Medium</priority>
          <bug_severity>normal</bug_severity>
          <target_milestone>1.5 M3</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Sebastian Schreiber">sschreiber</reporter>
          <assigned_to name="Nitin Kamble">nitin.a.kamble</assigned_to>
          <cc>bruce.ashfield</cc>
    
    <cc>dvhart</cc>
    
    <cc>tom.zanussi</cc>
          
          
          <cf_os>---</cf_os>
          <cf_regression_type>---</cf_regression_type>
          
          <cf_docchange>---</cf_docchange>

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>33946</commentid>
    <comment_count>0</comment_count>
    <who name="Sebastian Schreiber">sschreiber</who>
    <bug_when>2013-06-17 20:06:00 +0000</bug_when>
    <thetext>meta/recipes-kernel/linux/linux-yocto_3.8.bb invokes KERNEL_EXTRA_FEATURE netfilter.scc: compared to kernel &lt; 3.8 iptables with nat is not working.
I assume that CONFIG_NF_NAT_IP4 , CONFIG_NF_TARGET_MASQUERADE/NETMAP and REDIRECT should be set as Module in netfilter.cfg</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>34013</commentid>
    <comment_count>1</comment_count>
    <who name="Darren Hart">dvhart</who>
    <bug_when>2013-06-20 14:44:58 +0000</bug_when>
    <thetext>Sebastian, can you provide an example of how it is failing? Do you have a testcase?

I suspect you are correct and the meta-data needs to be reviewed. Would you care to test your suggestion below and submit a patch to the linux-yocto meta branch?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>34057</commentid>
    <comment_count>2</comment_count>
    <who name="Sebastian Schreiber">sschreiber</who>
    <bug_when>2013-06-22 15:44:01 +0000</bug_when>
    <thetext>Hi,
as a testcase you can u can build core-image-minimal for qemux86 with linux-yocto 3.8 and feature &apos;netfilter&apos;.
In the running system &apos;modprobe ip_tables&apos; is ok.
Using iptable with nat for e.g. forwarding fails with following return message:

 &gt;iptables -t nat -A POSTROUTING -o eth1 -j MASQUERADE
 &apos;iptables: Table does not exist (do you need to insmod?)&apos;

This doesn&apos;t happen with Kernel 3.4 or 3.2.

I&apos;ll provide a patch here later, as soon as i tested it.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>34058</commentid>
    <comment_count>3</comment_count>
      <attachid>1288</attachid>
    <who name="Sebastian Schreiber">sschreiber</who>
    <bug_when>2013-06-23 12:02:08 +0000</bug_when>
    <thetext>Created attachment 1288
netfilter kernel configuration fix for nat table

attached is a patch for meta feature that solves the problem. not sure if a patch for ip configuration is also needed.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>34934</commentid>
    <comment_count>4</comment_count>
    <who name="Nitin Kamble">nitin.a.kamble</who>
    <bug_when>2013-07-25 23:27:23 +0000</bug_when>
    <thetext>I could reproduce the issue locally. Will try the suggested fix now.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>35016</commentid>
    <comment_count>5</comment_count>
    <who name="Nitin Kamble">nitin.a.kamble</who>
    <bug_when>2013-07-26 21:16:38 +0000</bug_when>
    <thetext>The issue gets fixed with this extra addition to local.conf

KERNEL_FEATURES_sugarbay += &quot;features/netfilter/netfilter&quot;

Also the /proc/config.gz shows as expected. 

So this is not a kernel config issue, but a feature enabling matter. By default the netfilter kernel feature is not enabled for sugarbay.

So the right fix would be to enable the netfilter kernel feature.

Is it required to enable the netfilter feature in general for the sugarbay BSP, and other BSPs too?

It is enabled for qemu BSPs.

Nitin</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>35017</commentid>
    <comment_count>6</comment_count>
    <who name="Darren Hart">dvhart</who>
    <bug_when>2013-07-26 21:28:25 +0000</bug_when>
    <thetext>The question we have to ask is why did this work for 3.4 and stop working for 3.8?

Did we intentionally drop netfilter support from the standard ktype?

This strikes me as a regression if it is indeed true that netfilter was enabled by default previously. If so, the netfilter feature should probably be added to standard.scc.

Can you confirm that support was indeed dropped from 3.4 to 3.8?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>35018</commentid>
    <comment_count>7</comment_count>
    <who name="Nitin Kamble">nitin.a.kamble</who>
    <bug_when>2013-07-27 01:09:35 +0000</bug_when>
    <thetext>The problem is KERNEL_FEATURES additions in the meta-intel layer are overriding the previous value from oecore layer.


# $KERNEL_FEATURES [10 operations]
#   _append /home/nitin/build-test-bsps/poky.git/meta/recipes-kernel/linux/linux-yocto_3.8.bb:28
#     &quot; ${KERNEL_EXTRA_FEATURES}&quot;
#   _append[qemux86] /home/nitin/build-test-bsps/poky.git/meta/recipes-kernel/linux/linux-yocto_3.8.bb:29
#     &quot; cfg/sound.scc cfg/paravirt_kvm.scc&quot;
#   _append[qemux86-64] /home/nitin/build-test-bsps/poky.git/meta/recipes-kernel/linux/linux-yocto_3.8.bb:30
#     &quot; cfg/sound.scc&quot;
#   _append /home/nitin/build-test-bsps/poky.git/meta/recipes-kernel/linux/linux-yocto_3.8.bb:31
#     &quot; ${@bb.utils.contains(&quot;TUNE_FEATURES&quot;, &quot;mx32&quot;, &quot; cfg/x32.scc&quot;, &quot;&quot; ,d)}&quot;
#   set data_smart.py:414 [finalize]
#     &quot; ${KERNEL_EXTRA_FEATURES}&quot;
#   set data_smart.py:414 [finalize]
#     &quot; ${KERNEL_EXTRA_FEATURES} ${@bb.utils.contains(&quot;TUNE_FEATURES&quot;, &quot;mx32&quot;, &quot; cfg/x32.scc&quot;, &quot;&quot; ,d)}&quot;
#   finalize ast.py:333 [finalize]
#     &quot;was:  ${KERNEL_EXTRA_FEATURES} ${@bb.utils.contains(&quot;TUNE_FEATURES&quot;, &quot;mx32&quot;, &quot; cfg/x32.scc&quot;, &quot;&quot; ,d)}&quot;
#   override[sugarbay]:_append /home/nitin/build-test-bsps/poky.git/../meta-intel.git/meta-sugarbay/recipes-kernel/linux/linux-yocto_3.8.bbappend:7
#     &quot; features/usb/usb-uvcvideo features/media/v4l2&quot;
#   override[sugarbay]:set data_smart.py:414 [finalize]
#     &quot; features/usb/usb-uvcvideo features/media/v4l2&quot;
#   finalize override[sugarbay]:
#     &quot; features/usb/usb-uvcvideo features/media/v4l2&quot;
# computed:
#   &quot; features/usb/usb-uvcvideo features/media/v4l2&quot;
KERNEL_FEATURES=&quot;features/usb/usb-uvcvideo features/media/v4l2&quot;</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>35019</commentid>
    <comment_count>8</comment_count>
    <who name="Nitin Kamble">nitin.a.kamble</who>
    <bug_when>2013-07-27 01:49:50 +0000</bug_when>
    <thetext>It was an unintentional disabling of kernel features for bunch of BSPs. I have fixes for all these affected BSPs now. I will send the change for review shortly.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>35130</commentid>
    <comment_count>9</comment_count>
    <who name="Nitin Kamble">nitin.a.kamble</who>
    <bug_when>2013-07-29 18:30:02 +0000</bug_when>
    <thetext>a fix for this is sent to meta-intel mailing list.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>35236</commentid>
    <comment_count>10</comment_count>
    <who name="Nitin Kamble">nitin.a.kamble</who>
    <bug_when>2013-07-31 19:52:35 +0000</bug_when>
    <thetext>Fixed in meta-intel:

commit 6d074cae27420c9148aab66fda535f89cb6b7ca0
Author: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
Date:   Fri Jul 26 18:31:33 2013 -0700

    fri2: correct the KERNEL_FEATURE variable name
    
    The var KERNEL_FEATURES_sugarbay_append overrides the previous value
    defined in the oecore layer,
    and the var KERNEL_FEATURES_append_sugarbay preserves the old value
    and extends it.
    
    This change avoids unintentional disabling of some KERNEL_FEATURES
    such as netfilter.
    
    Signed-off-by: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
    Signed-off-by: Tom Zanussi &lt;tom.zanussi@linux.intel.com&gt;

commit 84fdca145a19d4715e2d743dd7b45c20e8ed9691
Author: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
Date:   Fri Jul 26 18:30:51 2013 -0700

    emenlow: correct the KERNEL_FEATURE variable name
    
    The var KERNEL_FEATURES_sugarbay_append overrides the previous value
    defined in the oecore layer,
    and the var KERNEL_FEATURES_append_sugarbay preserves the old value
    and extends it.
    
    This change avoids unintentional disabling of some KERNEL_FEATURES
    such as netfilter.
    
    Signed-off-by: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
    Signed-off-by: Tom Zanussi &lt;tom.zanussi@linux.intel.com&gt;


commit 75ac6a2a98fe76653dfb55da5c5afd592ec18acb
Author: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
Date:   Fri Jul 26 18:30:04 2013 -0700

    crownbay: correct the KERNEL_FEATURE variable name
    
    The var KERNEL_FEATURES_sugarbay_append overrides the previous value
    defined in the oecore layer,
    and the var KERNEL_FEATURES_append_sugarbay preserves the old value
    and extends it.
    
    This change avoids unintentional disabling of some KERNEL_FEATURES
    such as netfilter.
    
    Signed-off-by: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
    Signed-off-by: Tom Zanussi &lt;tom.zanussi@linux.intel.com&gt;

commit 2173e8f778a381012028ee79d5dc3351606a9bdb
Author: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
Date:   Thu Jun 27 13:33:51 2013 -0700

    sugarbay: correct the KERNEL_FEATURE variable name
    
    The var KERNEL_FEATURES_sugarbay_append overrides the previous value
    defined in the oecore layer,
    and the var KERNEL_FEATURES_append_sugarbay preserves the old value
    and extends it.
    
    This change avoids unintentional disabling of some KERNEL_FEATURES
    such as netfilter.
    
    Fixes bug:
    [YOCTO #4741]
    
    Signed-off-by: Nitin A Kamble &lt;nitin.a.kamble@intel.com&gt;
    Signed-off-by: Tom Zanussi &lt;tom.zanussi@linux.intel.com&gt;</thetext>
  </long_desc>
      
          <attachment
              isobsolete="0"
              ispatch="1"
              isprivate="0"
          >
            <attachid>1288</attachid>
            <date>2013-06-23 12:02:08 +0000</date>
            <delta_ts>2013-06-23 12:02:08 +0000</delta_ts>
            <desc>netfilter kernel configuration fix for nat table</desc>
            <filename>netfilter.patch</filename>
            <type>text/plain</type>
            <size>894</size>
            <attacher name="Sebastian Schreiber">sschreiber</attacher>
            
              <data encoding="base64">LS0tIGxpbnV4L21ldGEvY2ZnL2tlcm5lbC1jYWNoZS9mZWF0dXJlcy9uZXRmaWx0ZXIvbmV0Zmls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</data>

          </attachment>
      

    </bug>

</bugzilla>