gmane.comp.sysutils.netcf.devel

[PATCH] Add bundled(gnulib) to RPM specfile

Laine Stump — 2012-05-15T15:26:29

According to Fedora guidelines, because we bundle gnulib we
need to add a virtual Provides: bundled(gnulib).

https://fedoraproject.org/wiki/Packaging:No_Bundled_Libraries#Requirement_if_you_bundle
---
 netcf.spec.in |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/netcf.spec.in b/netcf.spec.in
index cec69c1..234561b 100644
--- a/netcf.spec.in
+++ b/netcf.spec.in
< at >< at > -14,6 +14,8 < at >< at > BuildRequires:  libxml2-devel libxslt-devel
 BuildRequires:  libnl-devel
 Requires:       %{name}-libs = %{version}-%{release}
 
+Provides: bundled(gnulib)
+
 %description
 Netcf is a library used to modify the network configuration of a
 system. Network configurations are expressed in a platform-independent

Preparing for freebsd support pull request

Sean Bruno — 2012-05-11T00:07:25

So, Hiren and I are approaching our first milestone, where we think that
we're feature complete for freebsd.  We're working out a few details at
the moment, but its starting to look good.

In order for you guys to pull from us, what's the standard process for
us to follow such that you can review, accept(and/or reject) and pull
from our tree.

I assume there's some kind of "generate review-able patches" guidelines
we should follow to make your lives easy.

Sean


_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

netcf patches for debian (v2)

2012-04-28T01:24:08

Hi,

Thanks Eric and Laine for your earlier comments. Here is a small
set of 3 cleaner patches which are enough to make netcf build on
debian.   (patches on top of git head).  I will try my hand at
the autoconf magic needed for libnl support next.

thanks,
-serge

_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

[PATCH 0/9] Ubuntu netcf patches

Serge Hallyn — 2012-03-20T14:07:41

Hi guys,

following is the set of patches applied in the Ubuntu package.  I was
sure I had sent these before, but they're not in the archive.  I was
wondering why there had never been any discussion!  Sorry about that.

thanks,
-serge
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

Ubuntu patches for netcf

Laine Stump — 2012-03-19T18:27:16

Serge,

After getting some netcf-related libvirt bug reports identified as
coming from an ubuntu system, I realized that you must now have a
released netcf build for ubuntu, but it's based on an upstream netcf git
checkout rather than a release. Any chance you could post the patches
you made so that I can push them and make a new upstream release (I'll
probably bump it from 0.1.9 to 0.2.0 to celebrate a released working
port on a new platform :-)

Thanks!

Laine
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

change-* commands

hiren panchasara — 2012-03-15T07:24:30

    change-begin - mark the beginning of a set of revertible network config
changes
    change-commit - commit the pending network config changes
    change-rollback - rollback (revert) a set of network config changes

Hi folks,

Can someone please take some time out to explain how these commands are
supposed to work?
what is the input and output of these commands?
it seems we take snapshots of current configurations? can someone shed some
light, please?

Thanks in advance,
Hiren
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

How to determine "start mode"?

hiren panchasara — 2012-03-12T00:16:49

<start mode="none"/> or <start mode="onboot"/>


How is it decided?

Thanks,
Hiren
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

oh hey ... just a note from over here in freebsd.orgland

Sean Bruno — 2012-03-07T17:53:24

Just a note of introduction as Hiren has been nice enough to keep the
questions to you guys coming when we are on the bus in the morning to
the offices.

He and I are thwacking on netcf at the moment in order to move on to
libvirt and then on to virtualization support on FreeBSD.  So, you fine
folks are the first stop on this little journey that ends up with
libvirt support for BHyVe ... at some point.

Cheers.

Sean
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

define and undefine

hiren panchasara — 2012-03-06T07:31:32

"define" takes an xmlfile as an argument, is that same as output of dumpxml?

And undefine would delete the interface from the configuration altogether.

Is my understanding correct?

Thanks a lot,
Hiren
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

dumpxml output format of resulting xml file

hiren panchasara — 2012-02-28T00:50:57

Hi All,

I am looking inside /test/interface/ at ethernet-dhcp.xml and
ethernet-static.xml:

[1]ethernet-dhcp.xml

<interface type="ethernet" name="eth0">
  <start mode="none"/>
  <mac address="aa:bb:cc:dd:ee:ff"/>
  <mtu size="1492"/>
  <protocol family="ipv4">
    <dhcp/><!-- peerdns="no"/>-->
  </protocol>
</interface>

[2]ethernet-static.xml

<interface type="ethernet" name="eth1">
  <start mode="onboot"/>
  <protocol family="ipv4">
    <ip address="192.168.0.5" prefix="24"/>
    <route gateway="192.168.0.1"/>
  </protocol>
</interface>


What I understand is, when "dumpxml em0" is called, if em0 is an ethernet
interface whose address has been provided by dhcp, the o/p *must* look like
[1] and if address is statically allocated, o/p *must* look like [2].
Please correct me if that is not accurate.

I want to understand rationale behind this format. How are these decided.
For example,  why "mac address" and "mtu size" do not matter in [2]?

Thanks for all the work and help,
Hiren
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

Adding netcf support to FreeBSD

hiren panchasara — 2012-02-06T18:47:09

Hi Folks,

I am trying to make netcf work with FreeBSD.
I am looking for more documentation on xml part of it. (I think augeas is
used for xml/xslt?)

I can "list" the interfaces but "dumpxml" does not work as I need to
prepare xml/xsl scaffolding.

Any pointers would be appreciated.

Thanks,
Hiren
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

ncftool manpage

Serge Hallyn — 2011-12-20T16:44:29

I added nothing of value here, just turned the 'ncftool help' output
into a ncftool.1 manpage.

thanks,
-serge

.TH NCFTOOL 1 "December 20, 2011"
.SH NAME
ncftool \- Network configuration tool
.SH SYNOPSIS
.B ncftool [options]
.SH DESCRIPTION
ncftool is a command line utility to configure networking.
.SH OPTIONS
.TP
.B list [\-\-macs] [\-\-all] [\-\-inactive]
List the currently configured toplevel network interfaces

.br
.BR [\-\-macs]
show MAC addresses
.br
.BR [\-\-all]
show all (up & down) interfaces
.br
.BR [\-\-inactive]
show only inactive (down) interfaces
.TP
.B dumpxml [\-\-mac] [\-\-live] <name>
Dump the XML description of an interface

.br
.BR [\-\-mac]
interpret the name as MAC address
.br
.BR [\-\-life]
include information about the live interface
.br
.BR <name>
the name of the interface
.TP
.B define <xmlfile>
Define - define an interface from an XML file

.br
.BR <xmlfile>
file containing the XML description of the interface
.TP
.B undefine <iface>
Remove the configuration of an interface

.br
.BR <iface>
the name of the interface
.TP
.B ifup <iface>
Bring up an interface

.br
.BR <iface>
the name of the interface
.TP
.B ifdown <iface>
Bring down an interface

.br
.BR <iface>
the name of the interface
.TP
.B change-begin
Mark the beginning of a set of revertible network config changes
.TP
.B change-commit
Commit (makes permanent) of a set of network config changes
.TP
.B change-rollback
Rollback (revert) a set of network config changes
.TP
.B help [<command>]
List all commands or print details about one command

.br
.BR <command>
If specified, list help about the listed command.  Otherwise list all commands.
.TP
.B quit
Exit the program
.TP
.SH SEE ALSO
.BR interfaces (5)
.SH AUTHOR
Serge Hallyn <serge.hallyn< at >ubuntu.com>
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

copyrights

Serge E. Hallyn — 2011-11-17T18:08:47

Hi,

I'm trying to finish the debian/ubuntu netcf package, and trying to make
sure I get copyrights right.  The debian/copyright file I have so far is
below.  There are a lot of xml files under tests/ and data/ (i.e.
./tests/suse/schema/vlan.xml) which have no attribution.  Are they all
the same license and copyright as the main project, or were some taken
from other projects?

thanks,
-serge

====================== debian/copyright ==========================

Format: http://dep.debian.net/deps/dep5
Upstream-Name: netcf
Source: http://git.fedorahosted.org/git/?p=netcf.git

Files: tests/cutest.*
Copyright: 2003 Asim Jalis
License:
 .
 This software is provided 'as-is', without any express or implied
 warranty. In no event will the authors be held liable for any damages
 arising from the use of this software.
 .
 Permission is granted to anyone to use this software for any purpose,
 including commercial applications, and to alter it and redistribute it
 freely, subject to the following restrictions:
 .
 1. The origin of this software must not be misrepresented; you must not
 claim that you wrote the original software. If you use this software in
 a product, an acknowledgment in the product documentation would be
 appreciated but is not required.
 .
 2. Altered source versions must be plainly marked as such, and must not be
 misrepresented as being the original software.
 .
 3. This notice may not be removed or altered from any source distribution.

Files: *
Copyright: 2009-2011 David Lutterkort <lutter< at >redhat.com>
           2009-2011 Laine Stump <laine< at >laine.org>
           2009 Daniel Veillard <veillard< at >redhat.com>
           2009 Jonas Eriksson <jonas.j.eriksson< at >ericsson.com>
           2009,2011 Daniel Berrange <berrange< at >redhat.com>
           2010 Cole Robinson <crobinso< at >redhat.com>
           2010,2011 Eric Blake <eblake< at >redhat.com>
           2011 Osier Yang <jyang< at >redhat.com>
           2011 Aleix Conchillo <aleix< at >oblong.com>
License: GNU LESSER GENERAL PUBLIC LICENSE
 .
 This library is free software; you can redistribute it and/or
 modify it under the terms of the GNU Lesser General Public
 License as published by the Free Software Foundation; either
 version 2.1 of the License, or (at your option) any later version.
 .
 This library is distributed in the hope that it will be useful,
 but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 Lesser General Public License for more details.
 .
 You should have received a copy of the GNU Lesser General Public
 License along with this library; if not, write to the Free Software
 Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.

Files: debian/*
Copyright: 2011 Ubuntu <ubuntu< at >unknown>
License: GPL-2+
 This package is free software; you can redistribute it and/or modify
 it under the terms of the GNU General Public License as published by
 the Free Software Foundation; either version 2 of the License, or
 (at your option) any later version.
 .
 This package is distributed in the hope that it will be useful,
 but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 GNU General Public License for more details.
 .
 You should have received a copy of the GNU General Public License
 along with this program. If not, see <http://www.gnu.org/licenses/>
 .
 On Debian systems, the complete text of the GNU General
 Public License version 2 can be found in "/usr/share/common-licenses/GPL-2".

# Please also look if there are files or directories which have a
# different copyright/license attached and list them here.
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

Fix to allow definitions with multiple ipv4 addresses for arp monitoring

Jens Nyberg I — 2011-10-31T14:08:05

Hi!

It's me again and I have a new patch. I don't know if I went about this the correct way but it solved a problem we had.

We wanted to define multiple targets for arp monitoring. Having multiple targets actually works except it is disallowed by the parser so I created a new definition called ipv4-multi-addr to solve it.

This might not be the correct solution so please advice me on how to further improve this in order to better suit the netcf code base.

Thanks!

// Jens

Patch:

--- interface.rng       2010-06-17 15:40:45.000000000 +0200
+++ interface.rng.new   2011-10-31 14:56:36.000000000 +0100
< at >< at > -202,7 +202,7 < at >< at >
           </element>
           <element name="arpmon">
             <attribute name="interval"><ref name="uint"/></attribute>
-            <attribute name="target"><ref name="ipv4-addr"/></attribute>
+            <attribute name="target"><ref name="ipv4-multi-addr"/></attribute>
             <optional>
               <attribute name="validate">
                 <choice>
< at >< at > -446,6 +446,12 < at >< at >
     </data>
   </define>
 
+  <define name='ipv4-multi-addr'>
+    <data type='string'>
+      <param name="pattern">(((25[0-5])|(1[0-9]{2})|(2[0-4][0-9])|([0-9]{1,2}))\.){3}((25[0-5])|(1[0-9]{2})|(2[0-4][0-9])|([0-9]{1,2}))(,(((25[0-5])|(1[0-9]{2})|(2[0-4][0-9])|([0-9]{1,2}))\.){3}((25[0-5])|(1[0-9]{2})|(2[0-4][0-9])|([0-9]{1,2})))*</param>
+    </data>
+  </define>
+
   <define name='ipv4-prefix'>
     <data type='unsignedInt'>
       <param name="maxInclusive">32</param>
_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

Fixed a bug in netcf when netmask is bigger than 24

Jens Nyberg I — 2011-10-31T11:23:43

Hi,

I noticed a bug in netcf when setting a netmask larger than /24.

This patch consist of:

* Removed a -1 in the inet_ntop call which is the actual bug that made the string incorrect.
* Changed the way you calculate the netmask ip in a more intuitive way - feel free to ignore it if you want.

--- xslt_ext.c  2011-10-26 17:42:54.000000000 +0200
+++ xslt_ext.c.new      2011-10-26 17:39:16.000000000 +0200
< at >< at > -71,10 +71,10 < at >< at >
     struct in_addr netmask;
     xmlChar netmask_str[16];

-    netmask.s_addr = htonl(~((1 << (32 - prefix)) - 1));
+    netmask.s_addr = htonl(~(0xffffffff >> prefix));

     if (! inet_ntop(AF_INET, &netmask,
-                    (char *) netmask_str, sizeof(netmask_str) - 1)) {
+                    (char *) netmask_str, sizeof(netmask_str))) {
         xsltTransformError(xsltXPathGetTransformContext(ctxt), NULL, NULL,
                     "ipcalc:netmask: internal error: inet_ntop failed");
         goto error;


_______________________________________________
netcf-devel mailing list
netcf-devel< at >lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

debian version

Serge E. Hallyn — 2011-10-27T22:58:59

Hi,

for the last two days I've been trying to get the debian version of ncftool
(I think not much changes since I put it up at
http://people.canonical.com/~serge/netcf - I should put up a source tree, I
know) to actually write out a new /etc/network/interfaces files when doing
ncftool define br.xml.  It claims success, but /etc/network/interfaces never
gets written, even though aug_save() gets called.

I've instrumented augeas's aug_set() and tree_store_value() to verify that
the following are being set:

aug_set: setting /files/etc/network/interfaces/auto[3]/1 to br0
aug_set: setting /files/etc/network/interfaces/iface[3] to br0
aug_set: setting /files/etc/network/interfaces/iface[3]/family to inet
aug_set: setting /files/etc/network/interfaces/iface[3]/method to dhcp
aug_set: setting /files/etc/network/interfaces/iface[3]/mtu to 1500
aug_set: setting /files/etc/network/interfaces/iface[3]/bridge_ports to eth0
aug_set: setting /files/etc/network/interfaces/iface[3]/bridge_stp to off
aug_set: setting /files/etc/network/interfaces/iface[3]/bridge_fd to 0.01

tree_store_value() now looks as follows, in case you're wondering:

void tree_store_value(struct tree *tree, char **value) {
#if 0
    if (streqv(tree->value, *value)) {
        free(*value);
        *value = NULL;
        return;
    }
#endif
    if (tree->value != NULL) {
        free(tree->value);
        tree->value = NULL;
    }
    if (*value != NULL) {
        tree->value = *value;
        *value = NULL;
    }
    {   
        FILE *f = fopen("/tmp/log4", "a");
        if (f) {
            fprintf(f, "tree_set_value: setting %s to %s\n",
                tree->label, tree->value);
            fclose(f);
        }
    }
    tree_mark_dirty(tree);
}

and log4 does get populated with entries for each of the above (showing
only the last part of the path of course).  So it seems there should be
no doubt that the values are, in fact being set.

But when I do

    {  
        FILE *f = fopen("/tmp/log7", "a");
        aug_print(aug, f, NULL);
        fclose(f);
    }

either before or after the aug_save in drv_define, the new rules are not
there!  (no 'bridge_ports' to be found at all in the output, for instance)
So the rules are disappearing before we get to aug_save(0).

Any ideas?

-serge

PS - I've used augtool by hand to verify that doing

set /files/etc/network/interfaces/auto[3]/1 br0
set /files/etc/network/interfaces/iface[3] br0
set /files/etc/network/interfaces/iface[3]/family inet
set /files/etc/network/interfaces/iface[3]/method dhcp
set /files/etc/network/interfaces/iface[3]/mtu 1500
set /files/etc/network/interfaces/iface[3]/bridge_ports eth0
set /files/etc/network/interfaces/iface[3]/bridge_stp off
set /files/etc/network/interfaces/iface[3]/bridge_fd 0.01
save

creates the right new /etc/network/interfaces file.

PPS - here is the br.xml I'm using:
<interface type="bridge" name="br0">
  <start mode="onboot"/>
  <mtu size="1500"/>
  <protocol family="ipv4">
    <dhcp/>
  </protocol>
  <bridge stp="off" delay="0.01">
    <interface type="ethernet" name="eth0">
<!--      <mac address="ab:bb:cc:dd:ee:ff"/>-->
    </interface>
  </bridge>
</interface>

Not possible to define a netmask higher than 24through virsh

2011-10-21T12:08:41

virsh fails to create an interface that has a netmask higher than 24 bits

--
_______________________________________________
netcf-devel mailing list
netcf-devel-TuqUDEhatI7GMZAyRF5v151Ccm5ICvs9< at >public.gmane.org
https://fedorahosted.org/mailman/listinfo/netcf-devel

[PATCH] Fix build errors when using-Werror=unused-but-set-variable

Laine Stump — 2011-10-12T04:20:29

From: Laine Stump <laine-/P6jbUKV17TZ+VzJOa5vwg< at >public.gmane.org>

The following errors appeared when I began building on Fedora 16. In
all cases, the variables really are never used, so they've been
removed.

drv_redhat.c: In function 'find_ifcfg_path_by_device':
drv_redhat.c:169:20: error: variable 'aug' set but not used
                            [-Werror=unused-but-set-variable]
drv_redhat.c: In function 'list_interfaces':
drv_redhat.c:315:20: error: variable 'aug' set but not used
                            [-Werror=unused-but-set-variable]
drv_redhat.c: In function 'drv_lookup_by_name':
drv_redhat.c:465:20: error: variable 'aug' set but not used
                            [-Werror=unused-but-set-variable]
drv_redhat.c: In function 'aug_put_xml':
drv_redhat.c:555:13: error: variable 'toplevel' set but not used
                            [-Werror=unused-but-set-variable]
drv_redhat.c: In function 'drv_lookup_by_mac_string':
drv_redhat.c:942:20: error: variable 'aug' set but not used
                            [-Werror=unused-but-set-variable]
---
 src/drv_redhat.c |   17 ++++-------------
 1 files changed, 4 insertions(+), 13 deletions(-)

diff --git a/src/drv_redhat.c b/src/drv_redhat.c
index f290cdb..d2c7971 100644
--- a/src/drv_redhat.c
+++ b/src/drv_redhat.c
< at >< at > -166,11 +166,10 < at >< at > static char *find_ifcfg_path_by_hwaddr(struct netcf *ncf, const char *mac) {
  * interface by checking for an entry 'DEVICE=NAME'
  */
 static char *find_ifcfg_path_by_device(struct netcf *ncf, const char *name) {
-    struct augeas *aug = NULL;
     int ndevs = 0;
     char **devs = NULL;
 
-    aug = get_augeas(ncf);
+    get_augeas(ncf);
     ERR_BAIL(ncf);
 
     ndevs = aug_fmt_match(ncf, &devs, "%s[DEVICE = '%s']",
< at >< at > -312,9 +311,8 < at >< at > static int list_ifcfg_paths(struct netcf *ncf, char ***intf) {
 
 static int list_interfaces(struct netcf *ncf, char ***intf) {
     int nint = 0, result = 0;
-    struct augeas *aug = NULL;
 
-    aug = get_augeas(ncf);
+    get_augeas(ncf);
     ERR_BAIL(ncf);
 
     /* Look in augeas for all interfaces */
< at >< at > -462,9 +460,8 < at >< at > struct netcf_if *drv_lookup_by_name(struct netcf *ncf, const char *name) {
     struct netcf_if *nif = NULL;
     char *pathx = NULL;
     char *name_dup = NULL;
-    struct augeas *aug;
 
-    aug = get_augeas(ncf);
+    get_augeas(ncf);
     ERR_BAIL(ncf);
 
     pathx = find_ifcfg_path(ncf, name);
< at >< at > -552,17 +549,12 < at >< at > static int aug_put_xml(struct netcf *ncf, xmlDocPtr xml) {
                   EINTERNAL, "expected node labeled 'tree', not '%s'",
                   tree->name);
         path = xml_prop(tree, "path");
-        int toplevel = 1;
         /* This is a little drastic, since it clears out the file entirely */
         r = aug_rm(aug, path);
         ERR_THROW(r < 0, ncf, EINTERNAL, "aug_rm of '%s' failed", path);
         list_for_each(node, tree->children) {
             label = xml_prop(node, "label");
             value = xml_prop(node, "value");
-            /* We should mark the toplevel interface from the XSLT */
-            if (STREQ(label, "BRIDGE") || STREQ(label, "MASTER")) {
-                toplevel = 0;
-            }
             r = xasprintf(&lpath, "%s/%s", path, label);
             ERR_NOMEM(r < 0, ncf);
 
< at >< at > -939,7 +931,6 < at >< at > int drv_undefine(struct netcf_if *nif) {
 int drv_lookup_by_mac_string(struct netcf *ncf, const char *mac,
                              int maxifaces, struct netcf_if **ifaces)
 {
-    struct augeas *aug = NULL;
     char *path = NULL, *ifcfg = NULL;
     const char **names = NULL;
     int nmatches = 0;
< at >< at > -949,7 +940,7 < at >< at > int drv_lookup_by_mac_string(struct netcf *ncf, const char *mac,
 
     MEMZERO(ifaces, maxifaces);
 
-    aug = get_augeas(ncf);
+    get_augeas(ncf);
     ERR_BAIL(ncf);
 
     nmatches = aug_match_mac(ncf, mac, &matches);

[PATCH] ncftool: avoid memory leak

2011-09-27T06:25:06

From: Alex Jia <ajia-H+wXaHxf7aLQT0dZR+AlfA< at >public.gmane.org>

* src/ncftool.c: Calling allocation function 'ncf_lookup_by_name' to allocate
  memory to 'nif', however, it hasn't been released before return.

Signed-off-by: Alex Jia <ajia-H+wXaHxf7aLQT0dZR+AlfA< at >public.gmane.org>
---
 src/ncftool.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/src/ncftool.c b/src/ncftool.c
index c9e26a2..90c6cbf 100644
--- a/src/ncftool.c
+++ b/src/ncftool.c
< at >< at > -390,8 +390,10 < at >< at > static int cmd_undefine(const struct command *cmd) {
         return CMD_RES_ERR;
 
     r = ncf_if_undefine(nif);
-    if (r < 0)
+    if (r < 0) {
+        ncf_if_free(nif);
         return CMD_RES_ERR;
+    }
 
     printf("%s undefined\n", name);
     ncf_if_free(nif);

[PATCH] Fix missing vlan/bond/ethernet info indumpxml --live

Laine Stump — 2011-09-22T11:34:44

This patch addresses:

  https://bugzilla.redhat.com/show_bug.cgi?id=736920
  https://bugzilla.redhat.com/show_bug.cgi?id=739505

Commit 38f44cf (pushed between 0.1.8 and 0.1.9) put a check for an
invalid ifindex into the functions that add type-specific status data
to the xml document (e.g. the mac address of an ethernet, or tag of a
vlan). This check was inadvertently done incorrectly ("ifindex !=
RTNL_LINK_NOT_FOUND" rather than "ifindex == RTNL_LINK_NOT_FOUND").
This patch corrects that error.

Because these functions are a part of the code to gather info from the
kernel via libnl (netlink) calls, there is unfortunately no simple way
to incorporate a build-time test to prevent similar regressions in the
future; we will instead have to rely on separate testing.
---
 src/dutil_linux.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/dutil_linux.c b/src/dutil_linux.c
index 75492ba..730be45 100644
--- a/src/dutil_linux.c
+++ b/src/dutil_linux.c
< at >< at > -909,7 +909,7 < at >< at > static void add_ethernet_info(struct netcf *ncf,
     struct rtnl_link *filter_link = NULL;
 
     /* if interface isn't currently available, nothing to add */
-    if (ifindex != RTNL_LINK_NOT_FOUND)
+    if (ifindex == RTNL_LINK_NOT_FOUND)
         return;
 
     filter_link = rtnl_link_alloc();
< at >< at > -998,7 +998,7 < at >< at > static void add_vlan_info(struct netcf *ncf,
     struct rtnl_link *filter_link = NULL;
 
     /* if interface isn't currently available, nothing to add */
-    if (ifindex != RTNL_LINK_NOT_FOUND)
+    if (ifindex == RTNL_LINK_NOT_FOUND)
         return;
 
     filter_link = rtnl_link_alloc();
< at >< at > -1113,7 +1113,7 < at >< at > static void add_bond_info(struct netcf *ncf,
         = { doc, root, NULL, ifindex, ncf };
 
     /* if interface isn't currently available, nothing to add */
-    if (ifindex != RTNL_LINK_NOT_FOUND)
+    if (ifindex == RTNL_LINK_NOT_FOUND)
         return;
 
     nl_cache_foreach(ncf->driver->link_cache, add_bond_info_cb, &cb_data);

[PATCH 0/11] Support for Debian/Suse/Windows distros

Daniel P. Berrange — 2011-09-21T16:50:12

This is an update to

 https://fedorahosted.org/pipermail/netcf-devel/2011-September/000601.html

As before it includes the Debian driver which is fully functional.

In v2 though, it also includes

 - The Suse driver taken from OpenSuse netcf RPMs. This is only
   partially complete and apparently only works for plain ethernet
   devices, not bridging/bonding
 - The Windows driver developed by Adam Stokes. This is only
   partially complete, only able to list devices and bring them
   up / down. No XML query/define support at all.
 - Update to the bootstrap scripts
 - An automated build script

IMHO, we should include the Suse/Windows drivers as they are,
even though they're incomplete, since this might motivate
someone into fixing them up further