[PATCH] network: Fix interface enumeration on down

[Rodrigo A B Freire]

A "# service network stop" tries to stop already stopped or non-existent
interfaces when a 'tunnel' interface is present.
The function check_device_down will use tunnel0 as a reference interface
(when it is up) and the network script will try to stop every interface
existent in /etc/sysconfig/network-scripts/ifcfg-*, regardless if it actually
exists or is up.

This patch fixes the variable and check_device_down now checks the correct
interface and down/skip the interfaces accordingly.
Fixes: RHBZ 1315018

See attached patch.
 --- Rodrigo A B Freire .--. - ..--- .-. --- -..

[Lukáš Nykrýn]

Hello,

I don't think this patch is correct. I can have for example vlan
eth0.10 and in configuration in file ifcfg-myvlan, with this patch the
device would not be pulled down.

Regards
Lukas

Rodrigo A B Freire píše v So 05. 03. 2016 v 20:54 +0000:
> A "# service network stop" tries to stop already stopped or non
> -existent
> interfaces when a 'tunnel' interface is present.
> The function check_device_down will use tunnel0 as a reference
> interface
> (when it is up) and the network script will try to stop every
> interface
> existent in /etc/sysconfig/network-scripts/ifcfg-*, regardless if it
> actually
> exists or is up.
>
> This patch fixes the variable and check_device_down now checks the
> correct
> interface and down/skip the interfaces accordingly.
>
> Fixes: RHBZ 1315018
>
> See attached patch.
>
> --- Rodrigo A B Freire .--. - ..--- .-. --- -..
> _______________________________________________
> initscripts-devel mailing list
> [e-mail]
> https://lists.fedorahosted.org/admin/lists/[e-mail]
> dorahosted.org

[Lukáš Nykrýn]

This is totally untested guess, but I think that correct fix would be:

diff --git a/rc.d/init.d/network b/rc.d/init.d/network
index cbb9ddf..8cf851d 100755
--- a/rc.d/init.d/network
+++ b/rc.d/init.d/network
@@ -204,6 +204,7 @@ case "$1" in
        done

for i in $xdslinterfaces $bridgeinterfaces $vlaninterfaces
$remaining; do
+               unset DEVICE
                (. ./ifcfg-$i
                if [ -z "$DEVICE" ] ; then DEVICE="$i"; fi

Lukáš Nykrýn píše v Po 07. 03. 2016 v 12:35 +0100:
> Hello,
>
> I don't think this patch is correct. I can have for example vlan
> eth0.10 and in configuration in file ifcfg-myvlan, with this patch
> the
> device would not be pulled down.
>
> Regards
> Lukas
>
> Rodrigo A B Freire píše v So 05. 03. 2016 v 20:54 +0000:
> > A "# service network stop" tries to stop already stopped or non
> > -existent
> > interfaces when a 'tunnel' interface is present.
> > The function check_device_down will use tunnel0 as a reference
> > interface
> > (when it is up) and the network script will try to stop every
> > interface
> > existent in /etc/sysconfig/network-scripts/ifcfg-*, regardless if
> > it
> > actually
> > exists or is up.
> >
> > This patch fixes the variable and check_device_down now checks the
> > correct
> > interface and down/skip the interfaces accordingly.
> >
> > Fixes: RHBZ 1315018
> >
> > See attached patch.
> >
> > --- Rodrigo A B Freire .--. - ..--- .-. --- -..
> > _______________________________________________
> > initscripts-devel mailing list
> > [e-mail]
> > https://lists.fedorahosted.org/admin/lists/initscripts-devel@lists.
> > fe
> > dorahosted.org
>
> initscripts-devel mailing list
> [e-mail]
> https://lists.fedorahosted.org/admin/lists/[e-mail]
> dorahosted.org

[Rodrigo A B Freire]

Hi Lukás, thank you very much for your review.
You are right and spotted the problem, I was able to reproduce it. The change that you suggested works just nicely.
Please find attached a new patchset (also containing a tunnel interface ordering). --- Rodrigo A B Freire 
.--. - ..--- .-. --- -..

On Monday, March 7, 2016 9:15 AM, Lukáš Nykrýn  wrote:

This is totally untested guess, but I think that correct fix would be:

diff --git a/rc.d/init.d/network b/rc.d/init.d/network
index cbb9ddf..8cf851d 100755
--- a/rc.d/init.d/network
+++ b/rc.d/init.d/network
@@ -204,6 +204,7 @@ case "$1" in
        done
        
        for i in $xdslinterfaces $bridgeinterfaces $vlaninterfaces
$remaining; do
+              unset DEVICE
                (. ./ifcfg-$i
                if [ -z "$DEVICE" ] ; then DEVICE="$i"; fi

Lukáš Nykrýn píše v Po 07. 03. 2016 v 12:35 +0100:
> Hello,
>
> I don't think this patch is correct. I can have for example vlan
> eth0.10 and in configuration in file ifcfg-myvlan, with this patch
> the
> device would not be pulled down.
>
> Regards
> Lukas
>
> Rodrigo A B Freire píše v So 05. 03. 2016 v 20:54 +0000:
> > A "# service network stop" tries to stop already stopped or non
> > -existent
> > interfaces when a 'tunnel' interface is present.
> > The function check_device_down will use tunnel0 as a reference
> > interface
> > (when it is up) and the network script will try to stop every
> > interface
> > existent in /etc/sysconfig/network-scripts/ifcfg-*, regardless if
> > it
> > actually
> > exists or is up.
> >
> > This patch fixes the variable and check_device_down now checks the
> > correct
> > interface and down/skip the interfaces accordingly.
> >
> > Fixes: RHBZ 1315018
> >
> > See attached patch.
> >   
> > --- Rodrigo A B Freire .--. - ..--- .-. --- -..
> > _______________________________________________
> > initscripts-devel mailing list
> > [e-mail]
> > https://lists.fedorahosted.org/admin/lists/initscripts-devel@lists.
> > fe
> > dorahosted.org
>
> initscripts-devel mailing list
> [e-mail]
> https://lists.fedorahosted.org/admin/lists/[e-mail]
> dorahosted.org

_______________________________________________
initscripts-devel mailing list
[e-mail]
https://lists.fedorahosted.org/admin/lists/[e-mail]...

[Lukáš Nykrýn]

Looks good to me. Unfortunately it is late for 6.8. So lets move this
to 6.9 "merge window".

Lukas

Rodrigo A B Freire píše v Po 07. 03. 2016 v 15:43 +0000:
> Hi Lukás, thank you very much for your review.
>
> You are right and spotted the problem, I was able to reproduce it.
> The change that you suggested works just nicely.
>
> Please find attached a new patchset (also containing a tunnel
> interface ordering).
>
> --- Rodrigo A B Freire 
> .--. - ..--- .-. --- -..
>
> On Monday, March 7, 2016 9:15 AM, Lukáš Nykrýn 
> wrote:
>
> This is totally untested guess, but I think that correct fix would
> be:
>
> diff --git a/rc.d/init.d/network b/rc.d/init.d/network
> index cbb9ddf..8cf851d 100755
> --- a/rc.d/init.d/network
> +++ b/rc.d/init.d/network
> @@ -204,6 +204,7 @@ case "$1" in
>         done
>     for i in $xdslinterfaces $bridgeinterfaces $vlaninterfaces
>
> $remaining; do
>          unset DEVICE
>           (. ./ifcfg-$i
>           if [ -z "$DEVICE" ] ; then DEVICE="$i"; fi
>
>
>
> Lukáš Nykrýn píše v Po 07. 03. 2016 v 12:35 +0100:
> > Hello,
> >
> > I don't think this patch is correct. I can have for example vlan
> > eth0.10 and in configuration in file ifcfg-myvlan, with this patch
> > the
> > device would not be pulled down.
> >
> > Regards
> > Lukas
> >
> > Rodrigo A B Freire píše v So 05. 03. 2016 v 20:54 +0000:
> > > A "# service network stop" tries to stop already stopped or non
> > > -existent
> > > interfaces when a 'tunnel' interface is present.
> > > The function check_device_down will use tunnel0 as a reference
> > > interface
> > > (when it is up) and the network script will try to stop every
> > > interface
> > > existent in /etc/sysconfig/network-scripts/ifcfg-*, regardless if
> > > it
> > > actually
> > > exists or is up.
> > >
> > > This patch fixes the variable and check_device_down now checks
>
> the
> > > correct
> > > interface and down/skip the interfaces accordingly.
> > >
> > > Fixes: RHBZ 1315018
> > >
> > > See attached patch.
> > >
> > > --- Rodrigo A B Freire .--. - ..--- .-. --- -..
> > > _______________________________________________
> > > initscripts-devel mailing list
> > > [e-mail]
>
> https://lists.fedorahosted.org/admin/lists/initscripts-devel@lists.
> > > fe
> > > dorahosted.org
> >
> > initscripts-devel mailing list
> > [e-mail]
>
> https://lists.fedorahosted.org/admin/lists/[e-mail]
> > dorahosted.org
>
> initscripts-devel mailing list
> [e-mail]
> https://lists.fedorahosted.org/admin/lists/[e-mail]
> dorahosted.org

[Rodrigo A B Freire]

Please deprecate the previous patch. As per your comment:
> Hello,
>
> I would like not to bring unnecessary differences between the rhel6 and
> master branch. So either we should use vpninterfaces in this case or
> introduce brifgeinterface variable in the patch for master branch.

I'm resending the RHEL6 patch renamed as vpninterfaces to be coherentwith rhel7/master. Thanks for pointing that. --- Rodrigo A B Freire .--. - ..--- .-. --- -..

On Saturday, March 5, 2016 5:54 PM, Rodrigo A B Freire  wrote:

A "# service network stop" tries to stop already stopped or non-existent
interfaces when a 'tunnel' interface is present.
The function check_device_down will use tunnel0 as a reference interface
(when it is up) and the network script will try to stop every interface
existent in /etc/sysconfig/network-scripts/ifcfg-*, regardless if it actually
exists or is up.

This patch fixes the variable and check_device_down now checks the correct
interface and down/skip the interfaces accordingly.
Fixes: RHBZ 1315018

See attached patch.
 --- Rodrigo A B Freire .--. - ..--- .-. --- -..